{ "version": 1, "updated": "2026-11-04", "description": "Slim index of the whitepaper monitor. A row with a catalog code (c) is cited in the principles catalog. Cited research (k = research) is not part of that default list. x = metadata means principles have not been extracted. Those rows are not sources of principles. ev is the monitor evidence tag, paraphrased on the page. possibly_deployed is the monitor tag tested, possibly deployed (needs review). It is not deployed.", "generated_from": [ "build/fixtures/whitepapers.json", "data/ai-agentic-principles.json" ], "fields": { "i": "WP- id", "t": "title", "o": "organization, from the monitor source_org field", "dt": "date published, as the monitor recorded it", "u": "URL", "n": "monitor issue number", "dec": "include, duplicate, or delete_candidate", "c": "catalog source codes for this monitor row, joined when two codes share a WP- id", "cs": "per-code catalog title, URL, organization, and date, in catalog order", "x": "extracted or metadata", "k": "org or research", "ev": "theoretical, tested, possibly_deployed, or deployed", "ax": "1 when the URL is on arxiv.org" }, "counts": { "cited": 68, "org": 105, "research": 400, "all": 579, "talks": 14, "principles": 170, "org-metadata": 3, "org-arxiv": 2, "research-extracted": 0, "ev-research-theoretical": 78, "ev-research-tested": 220, "ev-research-possibly_deployed": 102, "ev-research-deployed": 0, "ev-research-untagged": 0, "classic": 87, "sections": 22 }, "evidence_research": { "theoretical": 78, "tested": 220, "possibly_deployed": 102, "deployed": 0, "untagged": 0 }, "papers": [ { "i": "WP-0001", "t": "Security Engineering for Machine Learning (IEEE Computer)", "o": "Berryville Institute of Machine Learning (BIML)", "dt": "Aug 2019", "u": "https://apothecaryshed.files.wordpress.com/2019/08/biml-computer-8.19.pdf", "n": 1, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "BIML-SE", "ord": 20, "t": "Security Engineering for Machine Learning (IEEE Computer)", "u": "https://apothecaryshed.files.wordpress.com/2019/08/biml-computer-8.19.pdf", "o": "Berryville Institute of Machine Learning", "dt": "Aug 2019" } ], "c": "BIML-SE" }, { "i": "WP-0002", "t": "An Architectural Risk Analysis of Machine Learning Systems (BIML-78)", "o": "Berryville Institute of Machine Learning (BIML)", "dt": "Jan 20, 2020", "u": "https://berryvilleiml.com/results/ara.pdf", "n": 2, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "BIML-ARA", "ord": 21, "t": "An Architectural Risk Analysis of Machine Learning Systems (BIML-78)", "u": "https://berryvilleiml.com/results/ara.pdf", "o": "Berryville Institute of Machine Learning", "dt": "Jan 20, 2020" } ], "c": "BIML-ARA" }, { "i": "WP-0003", "t": "An Architectural Risk Analysis of Large Language Models (BIML-LLM24)", "o": "Berryville Institute of Machine Learning (BIML)", "dt": "Jan 24, 2024", "u": "https://berryvilleiml.com/results/BIML-LLM24.pdf", "n": 3, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "BIML-LLM24", "ord": 22, "t": "An Architectural Risk Analysis of Large Language Models (BIML-LLM24)", "u": "https://berryvilleiml.com/results/BIML-LLM24.pdf", "o": "Berryville Institute of Machine Learning", "dt": "Jan 24, 2024" } ], "c": "BIML-LLM24" }, { "i": "WP-0004", "t": "23 Security Risks in Black-Box Large Language Model Foundation Models (IEEE Computer)", "o": "Berryville Institute of Machine Learning (BIML)", "dt": "April 2024", "u": "https://apothecaryshed.com/wp-content/uploads/2024/05/23_Security_Risks_in_Black-Box_Large_Language_Model_Foundation_Models.pdf", "n": 4, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "BIML-23RISKS", "ord": 23, "t": "23 Security Risks in Black-Box Large Language Model Foundation Models (IEEE Computer)", "u": "https://apothecaryshed.com/wp-content/uploads/2024/05/23_Security_Risks_in_Black-Box_Large_Language_Model_Foundation_Models.pdf", "o": "Berryville Institute of Machine Learning", "dt": "April 2024" } ], "c": "BIML-23RISKS" }, { "i": "WP-0005", "t": "No Security Meter for AI", "o": "Berryville Institute of Machine Learning (BIML)", "dt": "May 13, 2026", "u": "https://berryvilleiml.com/results/no-security-meter-ai.pdf", "n": 5, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "BIML-NOMETER", "ord": 24, "t": "No Security Meter for AI", "u": "https://berryvilleiml.com/results/no-security-meter-ai.pdf", "o": "Berryville Institute of Machine Learning", "dt": "May 13, 2026" } ], "c": "BIML-NOMETER" }, { "i": "WP-0006", "t": "Artificial Intelligence and Large Language Models (LLM) Companion Guide (CIS Controls v8.1)", "o": "Center for Internet Security (CIS)", "dt": "April 20, 2026", "u": "https://www.cisecurity.org/insights/white-papers/controls-v8-1-ai-llm-companion-guide", "n": 6, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "CIS-LLM", "ord": 25, "t": "Artificial Intelligence and Large Language Models (LLM) Companion Guide (CIS Controls v8.1)", "u": "https://www.cisecurity.org/insights/white-papers/controls-v8-1-ai-llm-companion-guide", "o": "Center for Internet Security", "dt": "April 20, 2026" } ], "c": "CIS-LLM" }, { "i": "WP-0007", "t": "Artificial Intelligence (AI) Agents Companion Guide (CIS Controls v8.1)", "o": "Center for Internet Security (CIS)", "dt": "April 20, 2026", "u": "https://www.cisecurity.org/insights/white-papers/controls-v8-1-ai-agents-companion-guide", "n": 7, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "CIS-AGENTS", "ord": 0, "t": "Artificial Intelligence (AI) Agents Companion Guide (CIS Controls v8.1)", "u": "https://www.cisecurity.org/insights/white-papers/controls-v8-1-ai-agents-companion-guide", "o": "Center for Internet Security", "dt": "April 20, 2026" } ], "c": "CIS-AGENTS" }, { "i": "WP-0008", "t": "Model Context Protocol (MCP) Companion Guide (CIS Controls v8.1)", "o": "Center for Internet Security (CIS)", "dt": "April 20, 2026", "u": "https://www.cisecurity.org/insights/white-papers/controls-v8-1-model-context-protocol-companion-guide", "n": 8, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "CIS-MCP", "ord": 1, "t": "Model Context Protocol (MCP) Companion Guide (CIS Controls v8.1)", "u": "https://www.cisecurity.org/insights/white-papers/controls-v8-1-model-context-protocol-companion-guide", "o": "Center for Internet Security", "dt": "April 20, 2026" } ], "c": "CIS-MCP" }, { "i": "WP-0009", "t": "AI Controls Matrix (AICM) v1", "o": "Cloud Security Alliance (CSA)", "dt": "July 10, 2025", "u": "https://cloudsecurityalliance.org/artifacts/ai-controls-matrix", "n": 9, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "CSA-AICM1", "ord": 26, "t": "AI Controls Matrix (AICM) v1", "u": "https://cloudsecurityalliance.org/artifacts/ai-controls-matrix", "o": "Cloud Security Alliance", "dt": "July 10, 2025" } ], "c": "CSA-AICM1" }, { "i": "WP-0010", "t": "AI Controls Matrix (AICM) v1.1", "o": "Cloud Security Alliance (CSA)", "dt": "June 22, 2026", "u": "https://cloudsecurityalliance.org/artifacts/ai-controls-matrix-v1-1", "n": 10, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "CSA-AICM11", "ord": 27, "t": "AI Controls Matrix (AICM) v1.1", "u": "https://cloudsecurityalliance.org/artifacts/ai-controls-matrix-v1-1", "o": "Cloud Security Alliance", "dt": "June 22, 2026" } ], "c": "CSA-AICM11" }, { "i": "WP-0011", "t": "The State of Cloud and AI Security 2025", "o": "Cloud Security Alliance (CSA)", "dt": "2025", "u": "https://cloudsecurityalliance.org/artifacts/the-state-of-cloud-and-ai-security-2025", "n": 11, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "CSA-STATE", "ord": 28, "t": "The State of Cloud and AI Security 2025", "u": "https://cloudsecurityalliance.org/artifacts/the-state-of-cloud-and-ai-security-2025", "o": "Cloud Security Alliance", "dt": "2025" } ], "c": "CSA-STATE" }, { "i": "WP-0012", "t": "Agentic AI Identity & Access Management: A New Approach", "o": "Cloud Security Alliance (CSA)", "dt": "2026", "u": "https://cloudsecurityalliance.org/artifacts/agentic-ai-identity-and-access-management-a-new-approach", "n": 12, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "CSA-IAM", "ord": 2, "t": "Agentic AI Identity & Access Management: A New Approach", "u": "https://cloudsecurityalliance.org/artifacts/agentic-ai-identity-and-access-management-a-new-approach", "o": "Cloud Security Alliance", "dt": "2026" } ], "c": "CSA-IAM" }, { "i": "WP-0013", "t": "Data Security within AI Environments", "o": "Cloud Security Alliance (CSA)", "dt": "2025/2026", "u": "https://cloudsecurityalliance.org/artifacts/data-security-within-ai-environments", "n": 13, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "CSA-DATA", "ord": 29, "t": "Data Security within AI Environments", "u": "https://cloudsecurityalliance.org/artifacts/data-security-within-ai-environments", "o": "Cloud Security Alliance", "dt": "2025/2026" } ], "c": "CSA-DATA" }, { "i": "WP-0014", "t": "AI Vulnerability: Security Program Guide for CISOs", "o": "Cloud Security Alliance (CSA)", "dt": "Sept/Oct 2025", "u": "https://labs.cloudsecurityalliance.org/mythos-ciso/", "n": 14, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "CSA-VULN", "ord": 30, "t": "AI Vulnerability: Security Program Guide for CISOs", "u": "https://labs.cloudsecurityalliance.org/mythos-ciso/", "o": "Cloud Security Alliance", "dt": "Sept/Oct 2025" } ], "c": "CSA-VULN" }, { "i": "WP-0015", "t": "Establish Risks and Controls for the AI Supply Chain", "o": "Coalition for Secure AI (CoSAI)", "dt": "2024 (v1)", "u": "https://github.com/cosai-oasis/ws1-supply-chain/blob/main/whitepapers/risks-and-controls-for-the-ai-supply-chain-v1.md", "n": 15, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "COSAI-SUPPLY", "ord": 31, "t": "Establish Risks and Controls for the AI Supply Chain", "u": "https://github.com/cosai-oasis/ws1-supply-chain/blob/main/whitepapers/risks-and-controls-for-the-ai-supply-chain-v1.md", "o": "Coalition for Secure AI", "dt": "2024 (v1)" } ], "c": "COSAI-SUPPLY" }, { "i": "WP-0016", "t": "Signing ML Artifacts: Building Towards Tamper-Proof ML Metadata Records", "o": "Coalition for Secure AI (CoSAI)", "dt": "Sept 29, 2025 (blog summary)", "u": "https://github.com/cosai-oasis/ws1-supply-chain/blob/main/whitepapers/signing-ml-artifacts.md", "n": 16, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "COSAI-SIGN", "ord": 32, "t": "Signing ML Artifacts: Building Towards Tamper-Proof ML Metadata Records", "u": "https://github.com/cosai-oasis/ws1-supply-chain/blob/main/whitepapers/signing-ml-artifacts.md", "o": "Coalition for Secure AI", "dt": "Sept 29, 2025" } ], "c": "COSAI-SIGN" }, { "i": "WP-0017", "t": "Preparing Defenders of AI Systems", "o": "Coalition for Secure AI (CoSAI)", "dt": "2024", "u": "https://github.com/cosai-oasis/ws2-defenders/blob/main/preparing-defenders-of-ai-systems.md", "n": 17, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "COSAI-DEFEND", "ord": 33, "t": "Preparing Defenders of AI Systems", "u": "https://github.com/cosai-oasis/ws2-defenders/blob/main/preparing-defenders-of-ai-systems.md", "o": "Coalition for Secure AI", "dt": "2024" } ], "c": "COSAI-DEFEND" }, { "i": "WP-0018", "t": "AI Incident Response Framework", "o": "Coalition for Secure AI (CoSAI)", "dt": "Oct 30, 2025", "u": "https://github.com/cosai-oasis/ws2-defenders/blob/main/incident-response/AI-Incident-Response.md", "n": 18, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "COSAI-IR", "ord": 34, "t": "AI Incident Response Framework", "u": "https://github.com/cosai-oasis/ws2-defenders/blob/main/incident-response/AI-Incident-Response.md", "o": "Coalition for Secure AI", "dt": "Oct 30, 2025" } ], "c": "COSAI-IR" }, { "i": "WP-0019", "t": "CoSAI Principles for Secure-by-Design Agentic Systems", "o": "Coalition for Secure AI (CoSAI)", "dt": "July 14, 2025 (TSC approval)", "u": "https://github.com/cosai-oasis/cosai-tsc/blob/main/security-principles-for-agentic-systems.md", "n": 19, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "COSAI-SBD", "ord": 3, "t": "CoSAI Principles for Secure-by-Design Agentic Systems", "u": "https://github.com/cosai-oasis/cosai-tsc/blob/main/security-principles-for-agentic-systems.md", "o": "Coalition for Secure AI", "dt": "July 14, 2025" } ], "c": "COSAI-SBD" }, { "i": "WP-0020", "t": "Model Context Protocol (MCP) Security", "o": "Coalition for Secure AI (CoSAI)", "dt": "Jan 27, 2026", "u": "https://github.com/cosai-oasis/ws4-secure-design-agentic-systems/blob/main/whitepapers/model-context-protocol-security.md", "n": 20, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "COSAI-MCP", "ord": 4, "t": "Model Context Protocol (MCP) Security", "u": "https://github.com/cosai-oasis/ws4-secure-design-agentic-systems/blob/main/whitepapers/model-context-protocol-security.md", "o": "Coalition for Secure AI", "dt": "Jan 27, 2026" } ], "c": "COSAI-MCP" }, { "i": "WP-0021", "t": "The Future of Agentic Security: From Chatbots to Autonomous Swarms", "o": "Coalition for Secure AI (CoSAI)", "dt": "March 16, 2026", "u": "https://www.coalitionforsecureai.org/wp-content/uploads/2026/03/the-future-of-agentic-security.pdf", "n": 21, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "COSAI-FUTURE", "ord": 5, "t": "The Future of Agentic Security: From Chatbots to Autonomous Swarms", "u": "https://www.coalitionforsecureai.org/wp-content/uploads/2026/03/the-future-of-agentic-security.pdf", "o": "Coalition for Secure AI", "dt": "March 16, 2026" } ], "c": "COSAI-FUTURE" }, { "i": "WP-0022", "t": "Agentic Identity and Access Management", "o": "Coalition for Secure AI (CoSAI)", "dt": "April 17, 2026", "u": "https://www.coalitionforsecureai.org/wp-content/uploads/2026/04/agentic-identity-and-access-control.pdf", "n": 22, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "COSAI-IAM", "ord": 6, "t": "Agentic Identity and Access Management", "u": "https://www.coalitionforsecureai.org/wp-content/uploads/2026/04/agentic-identity-and-access-control.pdf", "o": "Coalition for Secure AI", "dt": "April 17, 2026" } ], "c": "COSAI-IAM" }, { "i": "WP-0023", "t": "AI Risk Management Framework (AI RMF 1.0)", "o": "National Institute of Standards and Technology (NIST)", "dt": "Jan 26, 2023", "u": "https://www.nist.gov/itl/ai-risk-management-framework", "n": 23, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "NIST-RMF", "ord": 35, "t": "AI Risk Management Framework (AI RMF 1.0)", "u": "https://www.nist.gov/itl/ai-risk-management-framework", "o": "National Institute of Standards and Technology", "dt": "Jan 26, 2023" } ], "c": "NIST-RMF" }, { "i": "WP-0024", "t": "AI RMF Playbook", "o": "National Institute of Standards and Technology (NIST)", "dt": "March 30, 2023", "u": "https://airc.nist.gov/airmf-resources/playbook/", "n": 24, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "NIST-PLAYBOOK", "ord": 36, "t": "AI RMF Playbook", "u": "https://airc.nist.gov/airmf-resources/playbook/", "o": "National Institute of Standards and Technology", "dt": "March 30, 2023" } ], "c": "NIST-PLAYBOOK" }, { "i": "WP-0025", "t": "AI RMF: Generative Artificial Intelligence Profile (NIST AI 600-1)", "o": "National Institute of Standards and Technology (NIST)", "dt": "July 26, 2024", "u": "https://nvlpubs.nist.gov/nistpubs/ai/NIST.AI.600-1.pdf", "n": 25, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "NIST-GAI", "ord": 37, "t": "AI RMF: Generative Artificial Intelligence Profile (NIST AI 600-1)", "u": "https://nvlpubs.nist.gov/nistpubs/ai/NIST.AI.600-1.pdf", "o": "National Institute of Standards and Technology", "dt": "July 26, 2024" } ], "c": "NIST-GAI" }, { "i": "WP-0026", "t": "Adversarial Machine Learning: A Taxonomy and Terminology of Attacks and Mitigations (NIST AI 100-2)", "o": "National Institute of Standards and Technology (NIST)", "dt": "2024, updated 2025", "u": "https://csrc.nist.gov/pubs/ai/100/2/e2025/final", "n": 26, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "NIST-ADVML", "ord": 38, "t": "Adversarial Machine Learning: A Taxonomy and Terminology of Attacks and Mitigations (NIST AI 100-2)", "u": "https://csrc.nist.gov/pubs/ai/100/2/e2025/final", "o": "National Institute of Standards and Technology", "dt": "2024, updated 2025" } ], "c": "NIST-ADVML" }, { "i": "WP-0027", "t": "AI Exchange (owaspai.org) , AI Security Overview & Controls Guide", "o": "OWASP AI Exchange", "dt": "Living document, updated 2022-2026", "u": "https://owaspai.org/", "n": 27, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "OWASP-AIEXCHANGE", "ord": 39, "t": "AI Exchange (owaspai.org): AI Security Overview & Controls Guide", "u": "https://owaspai.org/", "o": "OWASP AI Exchange", "dt": "Living document, 2022-2026" } ], "c": "OWASP-AIEXCHANGE" }, { "i": "WP-0028", "t": "OWASP Top 10 for LLM Applications & Generative AI (2025)", "o": "OWASP GenAI Security Project", "dt": "2025", "u": "https://genai.owasp.org/llm-top-10/", "n": 28, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "OWASP-LLMTOP10", "ord": 40, "t": "OWASP Top 10 for LLM Applications & Generative AI (2025)", "u": "https://genai.owasp.org/llm-top-10/", "o": "OWASP GenAI Security Project", "dt": "2025" } ], "c": "OWASP-LLMTOP10" }, { "i": "WP-0029", "t": "GenAI Incident Response Guide 1.0", "o": "OWASP GenAI Security Project", "dt": "July 28, 2025", "u": "https://genai.owasp.org/resource/genai-incident-response-guide-1-0/", "n": 29, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "OWASP-IR", "ord": 41, "t": "GenAI Incident Response Guide 1.0", "u": "https://genai.owasp.org/resource/genai-incident-response-guide-1-0/", "o": "OWASP GenAI Security Project", "dt": "July 28, 2025" } ], "c": "OWASP-IR" }, { "i": "WP-0030", "t": "Solutions Reference Guide Q2/Q3'25", "o": "OWASP GenAI Security Project", "dt": "Nov 4, 2025", "u": "https://genai.owasp.org/resource/owasp-genai-security-project-solutions-reference-guide-q2_q325/", "n": 30, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "OWASP-SOLUTIONS", "ord": 42, "t": "Solutions Reference Guide Q2/Q3'25", "u": "https://genai.owasp.org/resource/owasp-genai-security-project-solutions-reference-guide-q2_q325/", "o": "OWASP GenAI Security Project", "dt": "Nov 4, 2025" } ], "c": "OWASP-SOLUTIONS" }, { "i": "WP-0031", "t": "CheatSheet - A Practical Guide for Securely Using Third-Party MCP Servers 1.0", "o": "OWASP GenAI Security Project", "dt": "", "u": "https://genai.owasp.org/resource/cheatsheet-a-practical-guide-for-securely-using-third-party-mcp-servers-1-0/", "n": 31, "dec": "duplicate", "x": "extracted", "k": "org" }, { "i": "WP-0032", "t": "CheatSheet - A Practical Guide for Securely Using Third-Party MCP Servers 1.0", "o": "OWASP GenAI Security Project", "dt": "Nov 4, 2025", "u": "https://genai.owasp.org/resource/cheatsheet-a-practical-guide-for-securely-using-third-party-mcp-servers-1-0/", "n": 32, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "OWASP-MCP-CHEAT", "ord": 7, "t": "CheatSheet: A Practical Guide for Securely Using Third-Party MCP Servers 1.0", "u": "https://genai.owasp.org/resource/cheatsheet-a-practical-guide-for-securely-using-third-party-mcp-servers-1-0/", "o": "OWASP GenAI Security Project", "dt": "Nov 4, 2025" } ], "c": "OWASP-MCP-CHEAT" }, { "i": "WP-0033", "t": "OWASP Top 10 for Agentic Applications 2026", "o": "OWASP GenAI Security Project", "dt": "Dec 9, 2025", "u": "https://genai.owasp.org/resource/owasp-top-10-for-agentic-applications-for-2026/", "n": 33, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "OWASP-TOP10", "ord": 8, "t": "OWASP Top 10 for Agentic Applications 2026", "u": "https://genai.owasp.org/resource/owasp-top-10-for-agentic-applications-for-2026/", "o": "OWASP GenAI Security Project", "dt": "Dec 9, 2025" } ], "c": "OWASP-TOP10" }, { "i": "WP-0034", "t": "Vendor Evaluation Criteria for AI Red Teaming Providers & Tooling v1.0", "o": "OWASP GenAI Security Project", "dt": "Feb 4, 2026", "u": "https://genai.owasp.org/resource/owasp-vendor-evaluation-criteria-for-ai-red-teaming-providers-tooling-v1-0/", "n": 34, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "OWASP-REDTEAM", "ord": 43, "t": "Vendor Evaluation Criteria for AI Red Teaming Providers & Tooling v1.0", "u": "https://genai.owasp.org/resource/owasp-vendor-evaluation-criteria-for-ai-red-teaming-providers-tooling-v1-0/", "o": "OWASP GenAI Security Project", "dt": "Feb 4, 2026" } ], "c": "OWASP-REDTEAM" }, { "i": "WP-0035", "t": "A Practical Guide for Secure MCP Server Development", "o": "OWASP GenAI Security Project", "dt": "Feb 16, 2026", "u": "https://genai.owasp.org/resource/a-practical-guide-for-secure-mcp-server-development/", "n": 35, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "OWASP-MCP-DEV", "ord": 9, "t": "A Practical Guide for Secure MCP Server Development", "u": "https://genai.owasp.org/resource/a-practical-guide-for-secure-mcp-server-development/", "o": "OWASP GenAI Security Project", "dt": "Feb 16, 2026" } ], "c": "OWASP-MCP-DEV" }, { "i": "WP-0036", "t": "OWASP GenAI Data Security Risks & Mitigations 2026", "o": "OWASP GenAI Security Project", "dt": "March 17, 2026", "u": "https://genai.owasp.org/resource/owasp-genai-data-security-risks-mitigations-2026/", "n": 36, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "OWASP-DATASEC", "ord": 44, "t": "OWASP GenAI Data Security Risks & Mitigations 2026", "u": "https://genai.owasp.org/resource/owasp-genai-data-security-risks-mitigations-2026/", "o": "OWASP GenAI Security Project", "dt": "March 17, 2026" } ], "c": "OWASP-DATASEC" }, { "i": "WP-0037", "t": "State of Agentic AI Security and Governance 2.01", "o": "OWASP GenAI Security Project", "dt": "June 1, 2026", "u": "https://genai.owasp.org/resource/state-of-agentic-ai-security-and-governance/", "n": 37, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "OWASP-STATE", "ord": 10, "t": "State of Agentic AI Security and Governance 2.01", "u": "https://genai.owasp.org/resource/state-of-agentic-ai-security-and-governance/", "o": "OWASP GenAI Security Project", "dt": "June 1, 2026" } ], "c": "OWASP-STATE" }, { "i": "WP-0038", "t": "Critical AI Security Guidelines v1.0", "o": "SANS Institute", "dt": "March 31, 2025", "u": "https://www.sans.org/mlp/critical-ai-security-guidelines", "n": 38, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "SANS-CAISG1", "ord": 45, "t": "Critical AI Security Guidelines v1.0", "u": "https://www.sans.org/mlp/critical-ai-security-guidelines", "o": "SANS Institute", "dt": "March 31, 2025" } ], "c": "SANS-CAISG1" }, { "i": "WP-0039", "t": "Critical AI Security Guidelines v1.1 (draft)", "o": "SANS Institute", "dt": "April 2025", "u": "https://github.com/sans-community/ai-guidelines", "n": 39, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "SANS-CAISG11", "ord": 46, "t": "Critical AI Security Guidelines v1.1 (draft)", "u": "https://github.com/sans-community/ai-guidelines", "o": "SANS Institute", "dt": "April 2025" } ], "c": "SANS-CAISG11" }, { "i": "WP-0040", "t": "Critical AI Security Guidelines v1.4", "o": "SANS Institute", "dt": "March 2026", "u": "https://assets.contentstack.io/v3/assets/bltabe50a4554f8e97f/blte964a6eef293d57e/whitepaper-critical-ai-security-guidelines", "n": 40, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "SANS-CAISG14", "ord": 48, "t": "Critical AI Security Guidelines v1.4", "u": "https://assets.contentstack.io/v3/assets/bltabe50a4554f8e97f/blte964a6eef293d57e/whitepaper-critical-ai-security-guidelines", "o": "SANS Institute", "dt": "March 2026" } ], "c": "SANS-CAISG14" }, { "i": "WP-0041", "t": "Own AI Securely: The SANS Secure AI Blueprint", "o": "SANS Institute", "dt": "2025", "u": "https://www.sans.org/white-papers/own-ai-securely-sans-secure-ai-blueprint", "n": 41, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "SANS-BLUEPRINT", "ord": 47, "t": "Own AI Securely: The SANS Secure AI Blueprint", "u": "https://www.sans.org/white-papers/own-ai-securely-sans-secure-ai-blueprint", "o": "SANS Institute", "dt": "2025" } ], "c": "SANS-BLUEPRINT" }, { "i": "WP-0042", "t": "AI Shared Responsibility Framework (SRF) v1.0", "o": "Coalition for Secure AI (CoSAI)", "dt": "May 26, 2026", "u": "https://www.coalitionforsecureai.org/wp-content/uploads/2026/05/CoSAI-Shared-Responsibility-Framework.pdf", "n": 42, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "COSAI-SRF", "ord": 11, "t": "AI Shared Responsibility Framework (SRF) v1.0", "u": "https://www.coalitionforsecureai.org/wp-content/uploads/2026/05/CoSAI-Shared-Responsibility-Framework.pdf", "o": "Coalition for Secure AI", "dt": "May 26, 2026" } ], "c": "COSAI-SRF" }, { "i": "WP-0043", "t": "OWASP GenAI LLM Top 10 2026", "o": "OWASP GenAI Security Project", "dt": "August 3, 2026", "u": "https://genai.owasp.org/resource/owasp-genai-llm-top-10-2026/", "n": 43, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "OWASP-LLMTOP10-2026", "ord": 50, "t": "OWASP GenAI LLM Top 10 2026", "u": "https://genai.owasp.org/resource/owasp-genai-llm-top-10-2026/", "o": "OWASP GenAI Security Project", "dt": "August 3, 2026" } ], "c": "OWASP-LLMTOP10-2026" }, { "i": "WP-0044", "t": "OWASP Artificial Intelligence Security Verification Standard (AISVS) 1.0", "o": "OWASP GenAI Security Project", "dt": "June 24, 2026", "u": "https://github.com/OWASP/AISVS/raw/main/1.0/dist/AISVS-1.0.pdf", "n": 44, "dec": "include", "x": "extracted", "k": "org" }, { "i": "WP-0045", "t": "OWASP Agentic Skills Top 10 (AST10)", "o": "OWASP GenAI Security Project", "dt": "2026", "u": "https://owasp.org/www-project-agentic-skills-top-10/", "n": 45, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "OWASP-AST10", "ord": 51, "t": "OWASP Agentic Skills Top 10 (AST10)", "u": "https://owasp.org/www-project-agentic-skills-top-10/", "o": "OWASP GenAI Security Project", "dt": "2026" } ], "c": "OWASP-AST10" }, { "i": "WP-0046", "t": "CIS Controls v8.1.2 AI Security Guidance Workbook", "o": "Center for Internet Security (CIS)", "dt": "July 27, 2026", "u": "https://www.cisecurity.org/insights/white-papers/controls-v8-1-2-ai-security-guidance-workbook", "n": 46, "dec": "include", "x": "extracted", "k": "org" }, { "i": "WP-0047", "t": "SANS AI Security Maturity Model eBook", "o": "SANS Institute", "dt": "May 11, 2026", "u": "https://www.sans.org/white-papers/sans-ai-security-maturity-model-ebook", "n": 47, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "SANS-MATURITY", "ord": 49, "t": "SANS AI Security Maturity Model eBook", "u": "https://www.sans.org/white-papers/sans-ai-security-maturity-model-ebook", "o": "SANS Institute", "dt": "May 11, 2026" } ], "c": "SANS-MATURITY" }, { "i": "WP-0048", "t": "AI Security Solutions Landscape for AI and Agentic Red Teaming Q2 2026", "o": "OWASP GenAI Security Project", "dt": "April 9, 2026", "u": "https://genai.owasp.org/resource/ai-security-solutions-landscape-for-ai-and-agentic-red-teaming-q2-2026/", "n": 48, "dec": "include", "x": "extracted", "k": "org" }, { "i": "WP-0049", "t": "Solutions Landscape - Red Teaming Taxonomy", "o": "OWASP GenAI Security Project", "dt": "June 28, 2026", "u": "https://genai.owasp.org/resource/solutions-landscape-red-teaming-taxonomy/", "n": 49, "dec": "include", "x": "extracted", "k": "org" }, { "i": "WP-0050", "t": "Model Context Protocol (MCP) Security V2.0", "o": "Coalition for Secure AI (CoSAI)", "dt": "", "u": "https://www.coalitionforsecureai.org/wp-content/uploads/2026/09/model-context-protocol-security-v2.0.pdf", "n": 50, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "COSAI-MCP2", "ord": 12, "t": "Model Context Protocol (MCP) Security V2.0", "u": "https://www.coalitionforsecureai.org/wp-content/uploads/2026/09/model-context-protocol-security-v2.0.pdf", "o": "Coalition for Secure AI", "dt": "September 2026" } ], "c": "COSAI-MCP2" }, { "i": "WP-0051", "t": "Addendum: Post-Quantum Cryptography Considerations for MCP", "o": "Coalition for Secure AI (CoSAI)", "dt": "", "u": "https://www.coalitionforsecureai.org/wp-content/uploads/2026/09/Addendum-MCP-and-PQC.pdf", "n": 51, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "COSAI-MCP-PQC", "ord": 14, "t": "Addendum: Post-Quantum Cryptography Considerations for MCP", "u": "https://www.coalitionforsecureai.org/wp-content/uploads/2026/09/Addendum-MCP-and-PQC.pdf", "o": "Coalition for Secure AI", "dt": "September 2026" } ], "c": "COSAI-MCP-PQC" }, { "i": "WP-0052", "t": "Zero Trust for AI Systems", "o": "Coalition for Secure AI (CoSAI)", "dt": "", "u": "https://www.coalitionforsecureai.org/wp-content/uploads/2026/09/Zero-Trust-for-AI-Systems.pdf", "n": 52, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "COSAI-ZT", "ord": 13, "t": "Zero Trust for AI Systems", "u": "https://www.coalitionforsecureai.org/wp-content/uploads/2026/09/Zero-Trust-for-AI-Systems.pdf", "o": "Coalition for Secure AI", "dt": "September 2026" } ], "c": "COSAI-ZT" }, { "i": "WP-0053", "t": "OWASP GenAI Security Project Unveils 2026 Top 10 for LLM Applications, New Agent Control Standard and Sponsors as Community Tops 30,000 Members", "o": "OWASP GenAI Security Project", "dt": "9/2/26", "u": "https://genai.owasp.org/2026/09/01/owasp-genai-security-project-unveils-2026-top-10-for-llm-applications-new-agent-control-standard-and-sponsors-as-community-tops-30000-members/", "n": 53, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0054", "t": "FinBot CTF Is Live: A Hands-On Companion to the OWASP GenAI Security Project", "o": "OWASP GenAI Security Project", "dt": "4/15/26", "u": "https://genai.owasp.org/2026/04/14/finbot-ctf-is-live-a-hands-on-companion-to-the-owasp-genai-security-project/", "n": 54, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0055", "t": "OWASP GenAI Exploit Round-up Report Q1 2026", "o": "OWASP GenAI Security Project", "dt": "4/15/26", "u": "https://genai.owasp.org/2026/04/14/owasp-genai-exploit-round-up-report-q1-2026/", "n": 55, "dec": "include", "x": "extracted", "k": "org" }, { "i": "WP-0056", "t": "OWASP GenAI Security Project Expands AI Security Frameworks Ahead of RSA 2026, Celebrates Continued Sponsor Support", "o": "OWASP GenAI Security Project", "dt": "3/18/26", "u": "https://genai.owasp.org/2026/03/17/owasp-genai-security-project-expands-ai-security-frameworks-ahead-of-rsa-2026-celebrates-continued-sponsor-support/", "n": 56, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0057", "t": "Evolving AI Transparency: The Journey of the AIBOM Generator and Its New Home at OWASP", "o": "OWASP GenAI Security Project", "dt": "12/18/25", "u": "https://genai.owasp.org/2025/12/18/evolving-ai-transparency-the-journey-of-the-aibom-generator-and-its-new-home-at-owasp/", "n": 57, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0058", "t": "OWASP Top 10 for Agentic Applications - The Benchmark for Agentic Security in the Age of Autonomous AI", "o": "OWASP GenAI Security Project", "dt": "12/10/25", "u": "https://genai.owasp.org/2025/12/09/owasp-top-10-for-agentic-applications-the-benchmark-for-agentic-security-in-the-age-of-autonomous-ai/", "n": 58, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0059", "t": "OWASP GenAI Security Project Releases Top 10 Risks and Mitigations for Agentic AI Security", "o": "OWASP GenAI Security Project", "dt": "12/10/25", "u": "https://genai.owasp.org/2025/12/09/owasp-genai-security-project-releases-top-10-risks-and-mitigations-for-agentic-ai-security/", "n": 59, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0060", "t": "OWASP Agentic AI Taxonomy in Action: From Theory to Tools", "o": "OWASP GenAI Security Project", "dt": "7/23/25", "u": "https://genai.owasp.org/2025/07/22/owasp-agentic-ai-taxonomy-in-action-from-theory-to-tools/", "n": 60, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0061", "t": "OWASP Gen AI Incident & Exploit Round-up, Q2’25", "o": "OWASP GenAI Security Project", "dt": "7/14/25", "u": "https://genai.owasp.org/2025/07/14/owasp-gen-ai-incident-exploit-round-up-q225/", "n": 61, "dec": "include", "x": "extracted", "k": "org" }, { "i": "WP-0062", "t": "SANS Names the Five Winners of Find Evil!, the Largest Practitioner Evaluation of Autonomous AI Incident Response Agents", "o": "SANS Institute", "dt": "", "u": "https://www.sans.org/press/announcements/sans-names-the-five-winners-of-find-evil-2026", "n": 62, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0063", "t": "AI Is the Second-Biggest Human Risk in the Workplace, SANS Institute's 2026 Security Awareness & Culture Report Finds", "o": "SANS Institute", "dt": "", "u": "https://www.sans.org/press/announcements/ai-second-biggest-human-risk-workplace-sans-institutes-2026-security-awareness-culture-report-finds", "n": 63, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0064", "t": "AI Use in Cybersecurity Jumped From 50% to 78% in a Year. AI-Related Failures Rose Sharply Too. New SANS Institute Survey Reveals a Governance Gap.", "o": "SANS Institute", "dt": "", "u": "https://www.sans.org/press/announcements/ai-use-cybersecurity-jumped-from-50-to-78-year-ai-related-failures-rose-sharply-too-new-sans-institute-survey-reveals-governance-gap", "n": 64, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0065", "t": "SANS Institute and Yokohama National University Institute of Advanced Sciences (IAS) Unite to Accelerate Cybersecurity Research and Talent in the Age of AI", "o": "SANS Institute", "dt": "", "u": "https://www.sans.org/press/announcements/sans-ias-unite-to-accelerate-cybersecurity-research-talent-age-of-ai", "n": 65, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0066", "t": "SANS Cyber Workforce Academy Opens Applications for Expanded Maryland Program, Adding ICS/OT and AI Security Tracks", "o": "SANS Institute", "dt": "", "u": "https://www.sans.org/press/announcements/sans-cyber-workforce-academy-opens-applications-expanded-maryland-program-adding-icsot-ai-security-tracks", "n": 66, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0067", "t": "SANS Institute Expands Cyber Leaders Podcast with New Episodes Tackling AI Risk, Geopolitical Conflict, and Frontline Defense", "o": "SANS Institute", "dt": "", "u": "https://www.sans.org/press/announcements/sans-institute-expands-cyber-leaders-podcast-with-new-episodes-tackling-ai-risk-geopolitical-conflict-frontline-defense", "n": 67, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0068", "t": "SANS Institute Releases AI Security Maturity Model to Close the Gap Between Enterprise AI Adoption and the Governance to Control It", "o": "SANS Institute", "dt": "", "u": "https://www.sans.org/press/announcements/ai-security-maturity-model-close-gap-between-enterprise-ai-adoption-governance", "n": 68, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0069", "t": "Global AI Security Standard Organizations Gather Under MOSAIC to Reduce Fragmentation", "o": "SANS Institute", "dt": "", "u": "https://www.sans.org/press/announcements/global-ai-security-standard-organizations-gather-mosaic-reduce-fragmentation", "n": 69, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0070", "t": "SANS Institute, Cloud Security Alliance, [un]prompted, and OWASP GenAI Security Project Release Emergency Strategy Briefing as AI-Driven Vulnerability Discovery Compresses Exploit Timelines from Weeks to Hours", "o": "SANS Institute", "dt": "", "u": "https://www.sans.org/press/announcements/emergency-strategy-briefing-ai-driven-vulnerability-discovery-compresses-exploit-timelines", "n": 70, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0071", "t": "SANS Research: The Cybersecurity Talent Shortage Narrative Is Wrong. The Real Crisis Is What Your Team Doesn't Know, Starting with AI", "o": "SANS Institute", "dt": "", "u": "https://www.sans.org/press/announcements/sans-research-cybersecurity-talent-shortage-narrative-wrong-real-crisis-what-your-team-doesnt-know-starting-ai", "n": 71, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0072", "t": "SANS and siberX Introduce NOW // AI to Help Leaders Manage AI Risk, Strengthen Security Controls, and Defend Enterprise Value", "o": "SANS Institute", "dt": "", "u": "https://www.sans.org/press/announcements/sans-siberx-introduce-now-ai-leaders-manage-ai-risk-strengthen-security-controls-defend-enterprise", "n": 72, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0073", "t": "SANS and siberX Introduce NOW // AI to Help Leaders Manage AI Risk, Strengthen Security Controls, and Defend Enterprise Value", "o": "SANS Institute", "dt": "", "u": "https://www.sans.org/press/announcements/sans-siberx-introduce-now-ai-leaders-manage-ai-risk-strengthen-security-controls-defend-enterprise", "n": 73, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0074", "t": "SANS Institute Launches Season 3 of “Cyber Leaders” Podcast, Featuring Unfiltered AI, Quantum, and Crisis Decision-Making Insights", "o": "SANS Institute", "dt": "", "u": "https://www.sans.org/press/announcements/sans-institute-launches-season-3-of-cyber-leaders-podcast", "n": 74, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0075", "t": "SANS Institute, AWS, Microsoft and Google Cloud Launch Cloud Security Exchange eBook to Tackle AI-Powered Threats", "o": "SANS Institute", "dt": "", "u": "https://www.sans.org/press/announcements/launch-cloud-security-exchange-ebook-tackle-ai-powered-threats", "n": 75, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0076", "t": "SANS Secure AI Blueprint Provides the Roadmap Leaders Need to Adopt AI While Minimizing Risk", "o": "SANS Institute", "dt": "", "u": "https://www.sans.org/press/announcements/secure-ai-blueprint-provides-roadmap-leaders-need", "n": 76, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0077", "t": "SANS Institute Launches Cloud Security Exchange 2025 to Equip Teams with Real-World Defenses for AI-Driven Threats", "o": "SANS Institute", "dt": "", "u": "https://www.sans.org/press/announcements/sans-launches-cloud-security-exchange-2025", "n": 77, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0078", "t": "Building a Safer AI Future: SANS Leads Coordinated Effort to Secure Artificial Intelligence", "o": "SANS Institute", "dt": "", "u": "https://www.sans.org/press/announcements/building-safer-ai-future-sans-leads-coordinated-effort-secure-artificial-intelligence", "n": 78, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0079", "t": "Advancing Threat Detection in 2025: AI, Automation, and the Future of Cybersecurity", "o": "SANS Institute", "dt": "", "u": "https://www.sans.org/press/announcements/advancing-threat-detection-2025-ai-automation-future-of-cybersecurity", "n": 79, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0080", "t": "SANS Institute Launches AI Cybersecurity Hackathon to Develop the First Generation of Open-Source AI Security Tools", "o": "SANS Institute", "dt": "", "u": "https://www.sans.org/press/announcements/sans-institute-launches-ai-cybersecurity-hackathon-develop-first-generation-open-source-ai-security-tools", "n": 80, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0081", "t": "SANS Institute Releases 2024 Detection & Response Survey Highlighting AI, Automation, and Cloud Security Challenges", "o": "SANS Institute", "dt": "", "u": "https://www.sans.org/press/announcements/sans-institute-releases-2024-detection-response-survey-highlighting-ai-automation-cloud-security-challenges", "n": 81, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0082", "t": "Building AI Security Confidence: SANS Unveils Toolkit to Guide Secure AI Adoption", "o": "SANS Institute", "dt": "", "u": "https://www.sans.org/press/announcements/building-ai-security-confidence-sans-unveils-toolkit-guide-secure-ai-adoption", "n": 82, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0083", "t": "Unlock the Future of Cybersecurity: Exclusive, Next Era AI Insights and Cutting-Edge Training at SANS Network Security 2024", "o": "SANS Institute", "dt": "", "u": "https://www.sans.org/press/announcements/unlock-future-cybersecurity-exclusive-next-era-ai-insights-cutting-edge-training-sans-network-security-2024", "n": 83, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0084", "t": "Bridging the Gender Gap in Cybersecurity: SANS Institute and Women in CyberSecurity (WiCyS) Unveil International Women’s Day Scholarship for Cutting-Edge AI Course", "o": "SANS Institute", "dt": "", "u": "https://www.sans.org/press/announcements/bridging-gender-gap-cybersecurity-sans-women-cybersecurity-unveil-international-womens-day-scholarship", "n": 84, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0085", "t": "Cybersecurity in Paradise: AI Lessons Dominate SANS Holiday Hack Challenge 2023", "o": "SANS Institute", "dt": "", "u": "https://www.sans.org/press/announcements/cybersecurity-paradise-ai-lessons-dominate-sans-holiday-hack-challenge-2023/", "n": 85, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0086", "t": "Beyond the AI Hype: GIAC's GMLE Certification Translates Theory to Tangible Skills", "o": "SANS Institute", "dt": "", "u": "https://www.sans.org/press/announcements/beyond-ai-hype-giacs-gmle-certification-translates-theory-tangible-skills/", "n": 86, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0087", "t": "Shadow AI Does Not Read Your Org Chart: Rethinking Identity Governance for Autonomous Agents", "o": "Cloud Security Alliance (CSA)", "dt": "9/3/26", "u": "https://cloudsecurityalliance.org/articles/shadow-ai-does-not-read-your-org-chart-rethinking-identity-governance-for-autonomous-agents", "n": 87, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "CSA-SHADOW", "ord": 17, "t": "Shadow AI Does Not Read Your Org Chart: Rethinking Identity Governance for Autonomous Agents", "u": "https://cloudsecurityalliance.org/articles/shadow-ai-does-not-read-your-org-chart-rethinking-identity-governance-for-autonomous-agents", "o": "Cloud Security Alliance", "dt": "September 3, 2026" } ], "c": "CSA-SHADOW" }, { "i": "WP-0088", "t": "The Human-Machine Partnership: Architectures for Reliable AI", "o": "Cloud Security Alliance (CSA)", "dt": "9/1/26", "u": "https://cloudsecurityalliance.org/articles/the-human-machine-partnership-architectures-for-reliable-ai", "n": 88, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0089", "t": "A Framework for AI Threat Readiness", "o": "Cloud Security Alliance (CSA)", "dt": "9/1/26", "u": "https://cloudsecurityalliance.org/articles/a-framework-for-ai-threat-readiness", "n": 89, "dec": "include", "x": "extracted", "k": "org" }, { "i": "WP-0090", "t": "What Zero Trust Can Teach Us About AI Watermarks", "o": "Cloud Security Alliance (CSA)", "dt": "8/27/26", "u": "https://cloudsecurityalliance.org/articles/what-zero-trust-can-teach-us-about-ai-watermarks", "n": 90, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0091", "t": "State of AI Cybersecurity 2026: 87% of Security Professionals are Seeing More AI-Driven Threats, But Few Feel Ready to Stop Them", "o": "Cloud Security Alliance (CSA)", "dt": "8/26/26", "u": "https://cloudsecurityalliance.org/articles/state-of-ai-cybersecurity-2026-87-of-security-professionals-are-seeing-more-ai-driven-threats-but-few-feel-ready-to-stop-them", "n": 91, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0092", "t": "Lessons Learned on Securing Multi-Agent Systems: NIST Agent Security RFI", "o": "Cloud Security Alliance (CSA)", "dt": "8/26/26", "u": "https://cloudsecurityalliance.org/articles/lessons-learned-on-securing-multi-agent-systems-nist-agent-security-rfi", "n": 92, "dec": "include", "x": "extracted", "k": "org" }, { "i": "WP-0093", "t": "Cloud Security 2026: When AI is the Weapon and the Target", "o": "Cloud Security Alliance (CSA)", "dt": "8/26/26", "u": "https://cloudsecurityalliance.org/articles/cloud-security-2026-when-ai-is-the-weapon-and-the-target", "n": 93, "dec": "include", "x": "extracted", "k": "org" }, { "i": "WP-0094", "t": "Enterprise Reality: Why Organizations Aren't as Prepared for AI Governance as They Think They Are", "o": "Cloud Security Alliance (CSA)", "dt": "8/26/26", "u": "https://cloudsecurityalliance.org/articles/enterprise-reality-why-organizations-aren-t-as-prepared-for-ai-governance-as-they-think-they-are", "n": 94, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0095", "t": "Beyond Deepfakes: Zero Trust Security for the AI Economy", "o": "Cloud Security Alliance (CSA)", "dt": "8/19/26", "u": "https://cloudsecurityalliance.org/articles/beyond-deepfakes-zero-trust-security-for-the-ai-economy", "n": 95, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0096", "t": "State of AI Cybersecurity 2026: 77% of Security Stacks Include AI, But Trust is Lagging", "o": "Cloud Security Alliance (CSA)", "dt": "8/18/26", "u": "https://cloudsecurityalliance.org/articles/state-of-ai-cybersecurity-2026-77-of-security-stacks-include-ai-but-trust-is-lagging", "n": 96, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0097", "t": "MITRE's New Framework: Securing the eBPF Layer Your AI Depends On", "o": "Cloud Security Alliance (CSA)", "dt": "8/18/26", "u": "https://cloudsecurityalliance.org/articles/mitre-s-new-framework-securing-the-ebpf-layer-your-ai-depends-on", "n": 97, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0098", "t": "MITRE's New Continuous Remote Attestation Framework for the AI Era", "o": "Cloud Security Alliance (CSA)", "dt": "8/18/26", "u": "https://cloudsecurityalliance.org/articles/mitre-s-new-continuous-remote-attestation-framework-for-the-ai-era", "n": 98, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0099", "t": "DW TV (Germany): Why OpenAI and Anthropic CEOs say AI development must slow down", "o": "Berryville Institute of Machine Learning (BIML)", "dt": "9/13/26", "u": "https://berryvilleiml.com/2026/09/13/dw-tv-germany-why-openai-and-anthropic-ceos-say-ai-development-must-slow-down/", "n": 99, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0100", "t": "Decipher podcast on the future of AI security", "o": "Berryville Institute of Machine Learning (BIML)", "dt": "9/3/26", "u": "https://berryvilleiml.com/2026/09/03/decipher-podcast-on-the-future-of-ai-security/", "n": 100, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0101", "t": "MCP Security, Version 2.0: From Threat Taxonomy to a Model You Can Actually Audit Against", "o": "Coalition for Secure AI (CoSAI)", "dt": "9/25/26", "u": "https://www.coalitionforsecureai.org/mcp-security-version-2-0-from-threat-taxonomy-to-a-model-you-can-actually-audit-against/", "n": 101, "dec": "duplicate", "x": "metadata", "k": "org" }, { "i": "WP-0102", "t": "Zero Trust for AI Systems: Why Authorization Can’t Live Inside the Model", "o": "Coalition for Secure AI (CoSAI)", "dt": "9/16/26", "u": "https://www.coalitionforsecureai.org/zero-trust-for-ai-systems-why-authorization-cant-live-inside-the-model/", "n": 102, "dec": "duplicate", "x": "metadata", "k": "org" }, { "i": "WP-0103", "t": "Treat Your Agent Like an Insider Threat: Why AI Sandboxing Can’t Wait", "o": "Coalition for Secure AI (CoSAI)", "dt": "8/25/26", "u": "https://www.coalitionforsecureai.org/treat-your-agent-like-an-insider-threat-why-ai-sandboxing-cant-wait/", "n": 103, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "COSAI-SANDBOX", "ord": 15, "t": "Treat Your Agent Like an Insider Threat: Why AI Sandboxing Can't Wait", "u": "https://www.coalitionforsecureai.org/treat-your-agent-like-an-insider-threat-why-ai-sandboxing-cant-wait/", "o": "Coalition for Secure AI", "dt": "August 25, 2026" } ], "c": "COSAI-SANDBOX" }, { "i": "WP-0104", "t": "CoSAI Year 2: The Future of Agentic Security", "o": "Coalition for Secure AI (CoSAI)", "dt": "7/20/26", "u": "https://www.coalitionforsecureai.org/cosai-year-2-the-future-of-agentic-security/", "n": 104, "dec": "duplicate", "x": "metadata", "k": "org" }, { "i": "WP-0105", "t": "Who’s Responsible When AI Goes Wrong? A New Framework Aims to Answer That Question", "o": "Coalition for Secure AI (CoSAI)", "dt": "5/28/26", "u": "https://www.coalitionforsecureai.org/whos-responsible-when-ai-goes-wrong-a-new-framework-aims-to-answer-that-question/", "n": 105, "dec": "duplicate", "x": "metadata", "k": "org" }, { "i": "WP-0106", "t": "Coalition for Secure AI Unveils New Agentic Identity and Security Research Following High-Profile Sessions at RSAC 2026", "o": "Coalition for Secure AI (CoSAI)", "dt": "5/6/26", "u": "https://www.coalitionforsecureai.org/coalition-for-secure-ai-unveils-new-agentic-identity-and-security-research-following-high-profile-sessions-at-rsac-2026/", "n": 106, "dec": "duplicate", "x": "metadata", "k": "org" }, { "i": "WP-0107", "t": "After RSAC™ 2026: The MCP Security Question Everyone Kept Asking", "o": "Coalition for Secure AI (CoSAI)", "dt": "4/29/26", "u": "https://www.coalitionforsecureai.org/after-rsac-2026-the-mcp-security-question-everyone-kept-asking/", "n": 107, "dec": "duplicate", "x": "metadata", "k": "org" }, { "i": "WP-0108", "t": "Who’s Minding the Agent? A New Framework for AI Identity and Access Control", "o": "Coalition for Secure AI (CoSAI)", "dt": "4/17/26", "u": "https://www.coalitionforsecureai.org/whos-minding-the-agent-a-new-framework-for-ai-identity-and-access-control/", "n": 108, "dec": "duplicate", "x": "metadata", "k": "org" }, { "i": "WP-0109", "t": "AIMM: Artifact Integrity Maturity Model for AI/ML Supply Chain Trust and Provenance", "o": "Coalition for Secure AI (CoSAI)", "dt": "September 7, 2026", "u": "https://github.com/cosai-oasis/ws1-supply-chain/blob/main/whitepapers/aimm-artifact-integrity-maturity-model.pdf", "n": 109, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "COSAI-AIMM", "ord": 53, "t": "AIMM: Artifact Integrity Maturity Model for AI/ML Supply Chain Trust and Provenance", "u": "https://github.com/cosai-oasis/ws1-supply-chain/blob/main/whitepapers/aimm-artifact-integrity-maturity-model.pdf", "o": "Coalition for Secure AI", "dt": "September 7, 2026" } ], "c": "COSAI-AIMM" }, { "i": "WP-0110", "t": "Agent Control Standard (ACS) v0.1.0", "o": "OWASP GenAI Security Project", "dt": "September 1, 2026", "u": "https://github.com/GenAI-Security-Project/agent-control-standard", "n": 110, "dec": "include", "x": "extracted", "k": "org" }, { "i": "WP-0111", "t": "GenAI Security Industry Framework Crosswalk v4.0.0", "o": "OWASP GenAI Security Project", "dt": "September 1, 2026", "u": "https://github.com/GenAI-Security-Project/crosswalk", "n": 111, "dec": "include", "x": "extracted", "k": "org" }, { "i": "WP-0112", "t": "Quantifying Agentic Misalignment in Multi-Agent Systems", "o": "SANS Institute", "dt": "October 1, 2026", "u": "https://www.sans.org/white-papers/quantifying-agentic-misalignment-multi-agent-systems", "n": 112, "dec": "include", "x": "extracted", "k": "org" }, { "i": "WP-0113", "t": "AIMM: Artifact Integrity Maturity Model for AI/ML Supply Chain Trust and Provenance", "o": "Coalition for Secure AI (CoSAI)", "dt": "", "u": "https://www.coalitionforsecureai.org/wp-content/uploads/2026/10/aimm-artifact-integrity-maturity-model.pdf", "n": 113, "dec": "duplicate", "x": "metadata", "k": "org" }, { "i": "WP-0114", "t": "CSA Welcomes NVIDIA Open Agent Safety Platform", "o": "Cloud Security Alliance (CSA)", "dt": "", "u": "https://cloudsecurityalliance.org/articles/cloud-security-alliance-welcomes-nvidia-open-agent-safety-platform-to-advance-its-mission-of-securing-the-agentic-control-plane", "n": 114, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0115", "t": "CSA Welcomes NVIDIA Open Agent Safety Platform", "o": "Cloud Security Alliance (CSA)", "dt": "September 28, 2026", "u": "https://cloudsecurityalliance.org/articles/cloud-security-alliance-welcomes-nvidia-open-agent-safety-platform-to-advance-its-mission-of-securing-the-agentic-control-plane", "n": 115, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0116", "t": "From Models to MCP Servers, Skills, and Plugins: Rethinking Trust in the AI Supply Chain", "o": "Cloud Security Alliance (CSA)", "dt": "September 24, 2026", "u": "https://cloudsecurityalliance.org/articles/from-models-to-mcp-servers-skills-and-plugins-rethinking-trust-in-the-ai-supply-chain", "n": 116, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0117", "t": "AI Raised the Cyber Stakes, But Best Practices Still Win", "o": "Center for Internet Security (CIS)", "dt": "", "u": "https://www.cisecurity.org/insights/blog/ai-raised-the-cyber-stakes-best-practices-still-win", "n": 117, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0118", "t": "How Much Assurance Does Your AI Supply Chain Need?", "o": "Coalition for Secure AI (CoSAI)", "dt": "October 5, 2026", "u": "https://www.coalitionforsecureai.org/how-much-assurance-does-your-ai-supply-chain-need/", "n": 118, "dec": "duplicate", "x": "metadata", "k": "org" }, { "i": "WP-0119", "t": "GenAI and Agentic AI Exploit Roundup Q3 2026", "o": "OWASP GenAI Security Project", "dt": "October 8, 2026", "u": "https://genai.owasp.org/2026/10/08/genai-and-agentic-ai-exploit-roundup-q3-2026/", "n": 119, "dec": "include", "x": "extracted", "k": "org" }, { "i": "WP-0120", "t": "Nation-State Cyber Threats, Critical Infrastructure Attacks and AI Risks to Dominate SANS CyberThreat 2026", "o": "SANS Institute", "dt": "", "u": "https://www.sans.org/press/announcements/nation-state-cyber-threats-critical-infrastructure-attacks-ai-risks-dominate", "n": 120, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0121", "t": "Idaho Murders Forensic Keynote and New Digital Forensics and Incident Response AI Frameworks Headline SANS DFIR Summit 2026", "o": "SANS Institute", "dt": "", "u": "https://www.sans.org/press/announcements/idaho-murders-forensic-keynote-and-new-digital-forensics-and-incident-response-ai-frameworks-headline-sans-dfir-summit-2026", "n": 121, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0122", "t": "Sanctioned Apps, Unsanctioned AI: Governing the Agents Inside Your SaaS Stack", "o": "Cloud Security Alliance (CSA)", "dt": "October 6, 2026", "u": "https://cloudsecurityalliance.org/articles/sanctioned-apps-unsanctioned-ai-governing-the-agents-inside-your-saas-stack", "n": 122, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0123", "t": "Securing Agentic AI with Zero Trust Microsegmentation", "o": "Cloud Security Alliance (CSA)", "dt": "September 29, 2026", "u": "https://cloudsecurityalliance.org/articles/securing-agentic-ai-with-zero-trust-microsegmentation", "n": 123, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "CSA-ZT-MICRO", "ord": 54, "t": "Securing Agentic AI with Zero Trust Microsegmentation", "u": "https://cloudsecurityalliance.org/articles/securing-agentic-ai-with-zero-trust-microsegmentation", "o": "Cloud Security Alliance", "dt": "September 29, 2026" } ], "c": "CSA-ZT-MICRO" }, { "i": "WP-0124", "t": "5 Ways AI is Changing Traditional Security Models According to Modern CISOs", "o": "Cloud Security Alliance (CSA)", "dt": "September 29, 2026", "u": "https://cloudsecurityalliance.org/articles/5-ways-ai-is-changing-traditional-security-models-according-to-modern-cisos", "n": 124, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0125", "t": "Building a Generative AI Layer into Legacy Business Applications", "o": "Cloud Security Alliance (CSA)", "dt": "September 29, 2026", "u": "https://cloudsecurityalliance.org/articles/building-a-generative-ai-layer-into-legacy-business-applications", "n": 125, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0126", "t": "4 Ways AI Is Changing the Public Safety Threat Landscape", "o": "Center for Internet Security (CIS)", "dt": "", "u": "https://www.cisecurity.org/insights/blog/4-ways-ai-changing-public-safety-threat-landscape", "n": 126, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0127", "t": "AI Agents - You Can’t Govern What You Can’t Discover", "o": "Cloud Security Alliance (CSA)", "dt": "October 14, 2026", "u": "https://cloudsecurityalliance.org/articles/ai-agents-you-can-t-govern-what-you-can-t-discover", "n": 127, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0128", "t": "I Love Google’s Agentic AI Security Roadmap: Adding MAESTRO Will Make It More Powerful , An Open Letter to Google", "o": "Cloud Security Alliance (CSA)", "dt": "October 13, 2026", "u": "https://cloudsecurityalliance.org/articles/i-love-google-s-agentic-ai-security-roadmap-adding-maestro-will-make-it-more-powerful-an-open-letter-to-google", "n": 128, "dec": "include", "x": "extracted", "k": "org" }, { "i": "WP-0129", "t": "Introducing The Builder Exchange: Inside Security Stories From the Companies Building Fastest With AI", "o": "Cloud Security Alliance (CSA)", "dt": "October 9, 2026", "u": "https://cloudsecurityalliance.org/articles/introducing-the-builder-exchange-inside-security-stories-from-the-companies-building-fastest-with-ai", "n": 129, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0130", "t": "The Agent Kill Switch is No Longer Optional", "o": "Cloud Security Alliance (CSA)", "dt": "October 8, 2026", "u": "https://cloudsecurityalliance.org/articles/the-agent-kill-switch-is-no-longer-optional", "n": 130, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0131", "t": "AI in Penetration Testing: What's Actually Working and What Isn't", "o": "Cloud Security Alliance (CSA)", "dt": "September 29, 2026", "u": "https://cloudsecurityalliance.org/articles/ai-in-penetration-testing-what-s-actually-working-and-what-isn-t", "n": 131, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0132", "t": "How to Secure AI in Stadium Operations", "o": "Cloud Security Alliance (CSA)", "dt": "September 29, 2026", "u": "https://cloudsecurityalliance.org/articles/how-to-secure-ai-in-stadium-operations", "n": 132, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0133", "t": "Industrial Scale AI Distillation is a Cloud Security Failure", "o": "Cloud Security Alliance (CSA)", "dt": "September 29, 2026", "u": "https://cloudsecurityalliance.org/articles/industrial-scale-ai-distillation-is-a-cloud-security-failure", "n": 133, "dec": "include", "x": "extracted", "k": "org" }, { "i": "WP-0134", "t": "GenAI and Agentic AI Exploit Roundup Q2 2026", "o": "OWASP GenAI Security Project", "dt": "July 9, 2026", "u": "https://genai.owasp.org/2026/07/08/genai-and-agentic-ai-exploit-roundup-q2-2026/", "n": 134, "dec": "include", "x": "extracted", "k": "org" }, { "i": "WP-0135", "t": "MCP Tool Exposure is a Privilege Grant: A Two-Gate Authorization Model for Agentic AI", "o": "Cloud Security Alliance (CSA)", "dt": "October 15, 2026", "u": "https://cloudsecurityalliance.org/articles/mcp-tool-exposure-is-a-privilege-grant-a-two-gate-authorization-model-for-agentic-ai", "n": 135, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "CSA-MCP-GATE", "ord": 16, "t": "MCP Tool Exposure is a Privilege Grant: A Two-Gate Authorization Model for Agentic AI", "u": "https://cloudsecurityalliance.org/articles/mcp-tool-exposure-is-a-privilege-grant-a-two-gate-authorization-model-for-agentic-ai", "o": "Cloud Security Alliance", "dt": "October 15, 2026" } ], "c": "CSA-MCP-GATE" }, { "i": "WP-0136", "t": "The OWASP Top 10 for LLM Applications (2026): What Changed and Why It Matters", "o": "Cloud Security Alliance (CSA)", "dt": "October 8, 2026", "u": "https://cloudsecurityalliance.org/articles/the-owasp-top-10-for-llm-applications-2026-what-changed-and-why-it-matters", "n": 136, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0137", "t": "Bounding the Frontier: Why Enterprise AI Needs Security Validation Before Scale", "o": "Cloud Security Alliance (CSA)", "dt": "October 8, 2026", "u": "https://cloudsecurityalliance.org/articles/bounding-the-frontier-why-enterprise-ai-needs-security-validation-before-scale", "n": 137, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0138", "t": "GDM AI Control Roadmap (v0.1)", "o": "Google DeepMind", "dt": "June 22, 2026", "u": "https://storage.googleapis.com/deepmind-media/DeepMind.com/Blog/securing-the-future-of-ai-agents/gdm-ai-control-roadmap.pdf", "n": 138, "dec": "include", "x": "extracted", "k": "org" }, { "i": "WP-0139", "t": "AA26-251A: China-Based Artificial Intelligence Companies Conducting Industrial-Scale Distillation Campaigns Against U.S. AI Companies", "o": "CISA", "dt": "September 8, 2026", "u": "https://www.cisa.gov/news-events/cybersecurity-advisories/aa26-251a", "n": 139, "dec": "include", "x": "extracted", "k": "org" }, { "i": "WP-0140", "t": "Zero Trust Microsegmentation Guidance", "o": "Cloud Security Alliance (CSA)", "dt": "September 8, 2026", "u": "https://cloudsecurityalliance.org/artifacts/zero-trust-microsegmentation-guidance", "n": 140, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "CSA-ZT-SEG", "ord": 55, "t": "Zero Trust Microsegmentation Guidance", "u": "https://cloudsecurityalliance.org/artifacts/zero-trust-microsegmentation-guidance", "o": "Cloud Security Alliance", "dt": "September 8, 2026" } ], "c": "CSA-ZT-SEG" }, { "i": "WP-0141", "t": "Digital Forensics & AI Investigations Framework", "o": "SANS Institute", "dt": "2026", "u": "https://assets.contentstack.io/v3/assets/bltabe50a4554f8e97f/blt310ceb203d90eccf/frameworks-digital-forensics", "n": 141, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "SANS-DFIR", "ord": 58, "t": "Digital Forensics & AI Investigations Framework", "u": "https://assets.contentstack.io/v3/assets/bltabe50a4554f8e97f/blt310ceb203d90eccf/frameworks-digital-forensics", "o": "SANS Institute", "dt": "2026" } ], "c": "SANS-DFIR" }, { "i": "WP-0142", "t": "Incident Response & AI Investigations Framework", "o": "SANS Institute", "dt": "2026", "u": "https://assets.contentstack.io/v3/assets/bltabe50a4554f8e97f/blt629b7cccb1305f62/framework-incident-response", "n": 142, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "SANS-IR-AI", "ord": 59, "t": "Incident Response & AI Investigations Framework", "u": "https://assets.contentstack.io/v3/assets/bltabe50a4554f8e97f/blt629b7cccb1305f62/framework-incident-response", "o": "SANS Institute", "dt": "2026" } ], "c": "SANS-IR-AI" }, { "i": "WP-0143", "t": "MITRE ATLAS (Adversarial Threat Landscape for AI Systems) v2026.09", "o": "MITRE", "dt": "", "u": "https://atlas.mitre.org/", "n": 143, "dec": "include", "x": "extracted", "k": "org" }, { "i": "WP-0144", "t": "ETSI EN 304 223 V2.1.1 , Securing Artificial Intelligence (SAI); Baseline Cyber Security Requirements for AI Models and Systems", "o": "ETSI", "dt": "", "u": "https://www.etsi.org/deliver/etsi_en/304200_304299/304223/02.01.01_60/en_304223v020101p.pdf", "n": 144, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "ETSI-304223", "ord": 61, "t": "ETSI EN 304 223 V2.1.1: Baseline Cyber Security Requirements for AI Models and Systems", "u": "https://www.etsi.org/deliver/etsi_en/304200_304299/304223/02.01.01_60/en_304223v020101p.pdf", "o": "ETSI", "dt": "December 2025" } ], "c": "ETSI-304223" }, { "i": "WP-0145", "t": "ENISA's view on Cybersecurity in the Frontier AI Era", "o": "ENISA", "dt": "", "u": "https://www.enisa.europa.eu/publications/enisas-view-on-cybersecurity-in-the-frontier-ai-era", "n": 145, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "ENISA-FRONTIER", "ord": 62, "t": "ENISA's view on Cybersecurity in the Frontier AI Era", "u": "https://www.enisa.europa.eu/publications/enisas-view-on-cybersecurity-in-the-frontier-ai-era", "o": "ENISA", "dt": "July 2026" } ], "c": "ENISA-FRONTIER" }, { "i": "WP-0146", "t": "ETSI TS 104 275 V1.1.1: Securing Artificial Intelligence (SAI); Application of Baseline Cyber Security Requirements for AI Models and Systems principles to Generative AI; Preventing Harm from Deepfakes", "o": "ETSI", "dt": "", "u": "https://www.etsi.org/deliver/etsi_ts/104200_104299/104275/01.01.01_60/ts_104275v010101p.pdf", "n": 146, "dec": "include", "x": "extracted", "k": "org" }, { "i": "WP-0147", "t": "Security-Focused Guide for AI Code Assistant Instructions", "o": "OpenSSF", "dt": "August 1, 2025", "u": "https://best.openssf.org/Security-Focused-Guide-for-AI-Code-Assistant-Instructions", "n": 147, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "OPENSSF-ASSIST", "ord": 65, "t": "Security-Focused Guide for AI Code Assistant Instructions", "u": "https://best.openssf.org/Security-Focused-Guide-for-AI-Code-Assistant-Instructions", "o": "OpenSSF", "dt": "August 1, 2025" } ], "c": "OPENSSF-ASSIST" }, { "i": "WP-0148", "t": "Emerging Security Practices for AI Agents", "o": "Frontier Model Forum", "dt": "", "u": "https://www.frontiermodelforum.org/issue-briefs/emerging-security-practices-for-ai-agents/", "n": 148, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "FMF-AGENTS", "ord": 63, "t": "Emerging Security Practices for AI Agents", "u": "https://www.frontiermodelforum.org/issue-briefs/emerging-security-practices-for-ai-agents/", "o": "Frontier Model Forum", "dt": "June 3, 2026" } ], "c": "FMF-AGENTS" }, { "i": "WP-0149", "t": "Managing Advanced Cyber Risks in Frontier AI Frameworks", "o": "Frontier Model Forum", "dt": "", "u": "https://www.frontiermodelforum.org/technical-reports/managing-advanced-cyber-risks-in-frontier-ai-frameworks/", "n": 149, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "FMF-CYBER", "ord": 64, "t": "Managing Advanced Cyber Risks in Frontier AI Frameworks", "u": "https://www.frontiermodelforum.org/technical-reports/managing-advanced-cyber-risks-in-frontier-ai-frameworks/", "o": "Frontier Model Forum", "dt": "February 13, 2026" } ], "c": "FMF-CYBER" }, { "i": "WP-0150", "t": "Guidance for Safe Foundation Model Deployment", "o": "Partnership on AI", "dt": "", "u": "https://partnershiponai.org/modeldeployment/", "n": 150, "dec": "include", "x": "extracted", "k": "org" }, { "i": "WP-0151", "t": "Model AI Governance Framework for Agentic AI v1.5", "o": "AI Verify Foundation", "dt": "", "u": "https://www.imda.gov.sg/assets/63438074-73f6-4dcc-a281-030f42642cf4.pdf", "n": 151, "dec": "duplicate", "x": "extracted", "k": "org" }, { "i": "WP-0152", "t": "Model AI Governance Framework for Agentic AI v1.5", "o": "AI Verify Foundation", "dt": "", "u": "https://www.imda.gov.sg/assets/63438074-73f6-4dcc-a281-030f42642cf4.pdf", "n": 152, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "AIVERIFY-MAGF", "ord": 66, "t": "Model AI Governance Framework for Agentic AI v1.5", "u": "https://www.imda.gov.sg/assets/63438074-73f6-4dcc-a281-030f42642cf4.pdf", "o": "AI Verify Foundation", "dt": "May 20, 2026" } ], "c": "AIVERIFY-MAGF" }, { "i": "WP-0153", "t": "SAF-MCP: Secure Agentic Framework for Model Context Protocol (formerly SAFE-MCP), Framework Model v2", "o": "OpenSSF", "dt": "", "u": "https://github.com/secure-agentic-framework/saf-mcp", "n": 153, "dec": "duplicate", "x": "extracted", "k": "org" }, { "i": "WP-0154", "t": "Secure Software Development Practices for Generative AI and Dual-Use Foundation Models: An SSDF Community Profile (NIST SP 800-218A)", "o": "National Institute of Standards and Technology (NIST)", "dt": "", "u": "https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-218A.pdf", "n": 161, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "NIST-218A", "ord": 52, "t": "Secure Software Development Practices for Generative AI and Dual-Use Foundation Models: An SSDF Community Profile (SP 800-218A)", "u": "https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-218A.pdf", "o": "National Institute of Standards and Technology", "dt": "July 2024" } ], "c": "NIST-218A" }, { "i": "WP-0155", "t": "Agents for Cyber Defense", "o": "Frontier Model Forum", "dt": "", "u": "https://www.frontiermodelforum.org/issue-briefs/agents-for-cyber-defense/", "n": 155, "dec": "duplicate", "x": "extracted", "k": "org" }, { "i": "WP-0156", "t": "CyberSafe Foundation and SANS Institute Launch AI Security Fellowship for Women in Cybersecurity Across Africa", "o": "SANS Institute", "dt": "", "u": "https://www.sans.org/press/announcements/cybersafe-sans-launch-ai-security-fellowship-women-cybersecurity-across-africa", "n": 156, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0157", "t": "AI Security is more than the model", "o": "Berryville Institute of Machine Learning (BIML)", "dt": "October 15, 2026", "u": "https://berryvilleiml.com/2026/10/14/ai-security-is-more-than-the-model/", "n": 157, "dec": "delete_candidate", "x": "metadata", "k": "org" }, { "i": "WP-0158", "t": "SAF-MCP: Secure Agentic Framework for Model Context Protocol (formerly SAFE-MCP), Framework Model v2", "o": "OpenSSF", "dt": "", "u": "https://github.com/secure-agentic-framework/saf-mcp", "n": 158, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "OPENSSF-SAF-MCP", "ord": 19, "t": "SAF-MCP: Secure Agentic Framework for Model Context Protocol (formerly SAFE-MCP), Framework Model v2", "u": "https://github.com/secure-agentic-framework/saf-mcp", "o": "OpenSSF", "dt": "2026" } ], "c": "OPENSSF-SAF-MCP" }, { "i": "WP-0159", "t": "Agent Privacy Risk Taxonomy V.01 (AILuminate Assessment Standard)", "o": "MLCommons", "dt": "", "u": "https://mlcommons.org/wp-content/uploads/2026/10/ML-Commons-Agent-Privacy-Taxonomy-v01.pdf", "n": 159, "dec": "include", "x": "extracted", "k": "org" }, { "i": "WP-0160", "t": "Agents for Cyber Defense", "o": "Frontier Model Forum", "dt": "", "u": "https://www.frontiermodelforum.org/issue-briefs/agents-for-cyber-defense/", "n": 160, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "FMF-DEFENSE", "ord": 67, "t": "Agents for Cyber Defense", "u": "https://www.frontiermodelforum.org/issue-briefs/agents-for-cyber-defense/", "o": "Frontier Model Forum", "dt": "2026" } ], "c": "FMF-DEFENSE" }, { "i": "WP-0161", "t": "MIT AI Risk Repository", "o": "Massachusetts Institute of Technology (MIT)", "dt": "Not checked", "u": "https://airisk.mit.edu/", "n": 162, "dec": "include", "x": "extracted", "k": "org" }, { "i": "WP-0162", "t": "MAESTRO: Agentic AI Threat Modeling Framework", "o": "Cloud Security Alliance (CSA)", "dt": "February 6, 2025 (checked on the CSA page)", "u": "https://cloudsecurityalliance.org/blog/2025/02/06/agentic-ai-threat-modeling-framework-maestro", "n": 163, "dec": "include", "x": "extracted", "k": "org" }, { "i": "WP-0163", "t": "NIST COSAIS: Control Overlays for Securing AI Systems", "o": "National Institute of Standards and Technology (NIST)", "dt": "August 14, 2025 (concept paper date, from the NIST project page). The project page was last updated October 20, 2026.", "u": "https://csrc.nist.gov/projects/cosais", "n": 164, "dec": "include", "x": "extracted", "k": "org" }, { "i": "WP-0164", "t": "Google Secure AI Framework (SAIF)", "o": "Google", "dt": "Not checked", "u": "https://saif.google/", "n": 165, "dec": "include", "x": "extracted", "k": "org" }, { "i": "WP-0165", "t": "An Introduction to Google's Approach for Secure AI Agents", "o": "Google", "dt": "May 2025 (from the paper PDF)", "u": "https://research.google/pubs/an-introduction-to-googles-approach-for-secure-ai-agents/", "n": 166, "dec": "include", "x": "extracted", "k": "org" }, { "i": "WP-0166", "t": "The Agentic AI Security Scoping Matrix", "o": "Amazon Web Services (AWS)", "dt": "November 21, 2025 (from the AWS page metadata)", "u": "https://aws.amazon.com/blogs/security/the-agentic-ai-security-scoping-matrix-a-framework-for-securing-autonomous-ai-systems/", "n": 167, "dec": "include", "x": "extracted", "k": "org" }, { "i": "WP-0167", "t": "Taxonomy of Failure Modes in AI Agents", "o": "Microsoft", "dt": "April 24, 2025 (from the URL; not checked on the page)", "u": "https://www.microsoft.com/en-us/security/blog/2025/04/24/new-whitepaper-outlines-the-taxonomy-of-failure-modes-in-ai-agents/", "n": 168, "dec": "include", "x": "extracted", "k": "org" }, { "i": "WP-0168", "t": "Agentic AI Security: Threats, Defenses, Evaluation, and Open Challenges (arXiv 2510.23883)", "o": "Academic survey by Anshuman Chhabra, Shrestha Datta, Shahriar Kabir Nahin, and Prasant Mohapatra (from the arXiv page)", "dt": "October 27, 2025 (arXiv v1). The arXiv comment says it was published in IEEE Access.", "u": "https://arxiv.org/abs/2510.23883", "n": 169, "dec": "include", "x": "extracted", "k": "org", "ev": "theoretical", "ax": 1 }, { "i": "WP-0169", "t": "TRiSM for Agentic AI: A Review of Trust, Risk, and Security Management in LLM-based Agentic Multi-Agent Systems (arXiv 2506.04133)", "o": "Academic review by Shaina Raza, Ranjan Sapkota, Manoj Karkee, and Christos Emmanouilidis (from the arXiv page)", "dt": "June 4, 2025 (arXiv v1; last revised December 18, 2025)", "u": "https://arxiv.org/abs/2506.04133", "n": 170, "dec": "include", "x": "extracted", "k": "org", "ev": "theoretical", "ax": 1 }, { "i": "WP-0170", "t": "MLCommons AILuminate Benchmark", "o": "MLCommons", "dt": "Not checked", "u": "https://mlcommons.org/benchmarks/ailuminate/", "n": 171, "dec": "include", "x": "extracted", "k": "org" }, { "i": "WP-0171", "t": "Agentic AI Top 10 Vulnerability (Precize)", "o": "Precize (GitHub project)", "dt": "Not checked", "u": "https://github.com/precize/Agentic-AI-Top10-Vulnerability", "n": 172, "dec": "include", "x": "extracted", "k": "org" }, { "i": "WP-0172", "t": "OWASP GenAI Red Teaming Manual Version 1.0", "o": "OWASP GenAI Security Project", "dt": "October 20, 2026 (resource page). The PDF cover says October 2026.", "u": "https://genai.owasp.org/resource/genai-red-teaming-manual-version-1-0/", "n": 174, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "OWASP-RT-MANUAL", "ord": 57, "t": "OWASP GenAI Red Teaming Manual Version 1.0", "u": "https://genai.owasp.org/resource/genai-red-teaming-manual-version-1-0/", "o": "OWASP GenAI Security Project", "dt": "October 2026" } ], "c": "OWASP-RT-MANUAL" }, { "i": "WP-0173", "t": "OWASP LLM Exploit Generation Phase 2: Agentic vs. Classic Offensive Approaches (LLM Exploit Generation Guide 2.0)", "o": "OWASP GenAI Security Project, Threat Intelligence Initiative", "dt": "October 27, 2026 (resource page). The PDF cover says Version 1.0, October 2026.", "u": "https://genai.owasp.org/resource/llm-exploit-generation-guide-2-0/", "n": 175, "dec": "include", "x": "extracted", "k": "org" }, { "i": "WP-0174", "t": "OWASP GenAI Data Security Best Practices v2, Part 1", "o": "OWASP GenAI Security Project", "dt": "November 3, 2026 (resource page). The PDF cover says Version 2.0, Part 1, October 2026.", "u": "https://genai.owasp.org/resource/genai-data-security-best-practices-v2-part1/", "n": 176, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "OWASP-DATASEC2", "ord": 56, "t": "OWASP GenAI Data Security Best Practices v2, Part 1", "u": "https://genai.owasp.org/resource/genai-data-security-best-practices-v2-part1/", "o": "OWASP GenAI Security Project", "dt": "October 2026" } ], "c": "OWASP-DATASEC2" }, { "i": "WP-0175", "t": "Design Tenets for Agent Identity Management Systems", "o": "National Institute of Standards and Technology (NIST), Cybersecurity Insights blog", "dt": "October 27, 2026 (from the NIST page)", "u": "https://www.nist.gov/blogs/cybersecurity-insights/design-tenets-agent-identity-management-systems", "n": 177, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "NIST-AGENT-ID", "ord": 18, "t": "Design Tenets for Agent Identity Management Systems", "u": "https://www.nist.gov/blogs/cybersecurity-insights/design-tenets-agent-identity-management-systems", "o": "National Institute of Standards and Technology", "dt": "October 27, 2026" } ], "c": "NIST-AGENT-ID" }, { "i": "WP-0176", "t": "Key Practices for Incident Readiness and Response in Super Intelligence Training and Evaluation", "o": "CISA, with NIST CAISSI and NSA SISC (as listed in the document)", "dt": "October 16, 2026 (CISA resource page). The PDF says October 2026, Version 1.0, PP-26-4396, TLP:CLEAR.", "u": "https://www.cisa.gov/resources-tools/resources/key-practices-incident-readiness-and-response-super-intelligence-training-and-evaluation", "n": 178, "dec": "include", "x": "extracted", "k": "org", "cs": [ { "c": "CISA-SI-IR", "ord": 60, "t": "Key Practices for Incident Readiness and Response in Super Intelligence Training and Evaluation", "u": "https://www.cisa.gov/resources-tools/resources/key-practices-incident-readiness-and-response-super-intelligence-training-and-evaluation", "o": "CISA", "dt": "October 2026" } ], "c": "CISA-SI-IR" }, { "i": "WP-0177", "t": "Frontier Ready Assessment Framework", "o": "Cloud Security Alliance (CSA)", "dt": "October 8, 2026 (CSA artifact page)", "u": "https://cloudsecurityalliance.org/artifacts/frontier-ready-assessment-framework", "n": 179, "dec": "include", "x": "metadata", "k": "org" }, { "i": "WP-0178", "t": "Transitioning from STAR to STAR for AI", "o": "Cloud Security Alliance (CSA)", "dt": "October 26, 2026 (CSA artifact page)", "u": "https://cloudsecurityalliance.org/artifacts/transitioning-from-star-to-star-for-ai", "n": 180, "dec": "include", "x": "metadata", "k": "org" }, { "i": "WP-0179", "t": "From Framework to Action: Applying the SANS AI Security Maturity Model", "o": "SANS Institute", "dt": "October 14, 2026 (SANS page)", "u": "https://www.sans.org/white-papers/from-framework-action-applying-sans-ai-security-maturity-model", "n": 181, "dec": "include", "x": "metadata", "k": "org" }, { "i": "WP-0180", "t": "Poisoning Attacks against Support Vector Machines (arXiv 1206.6389)", "o": "Academic paper by Battista Biggio, Blaine Nelson, Pavel Laskov (from the arXiv API)", "dt": "2012-06-27 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1206.6389", "n": 182, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0181", "t": "Hacking Smart Machines with Smarter Ones: How to Extract Meaningful Data from Machine Learning Classifiers (arXiv 1306.4447)", "o": "Academic paper by Giuseppe Ateniese, Giovanni Felici, Luigi V. Mancini, Angelo Spognardi, Antonio Villani, Domenico Vitali (from the arXiv API)", "dt": "2013-06-19 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1306.4447", "n": 183, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0182", "t": "Intriguing properties of neural networks (arXiv 1312.6199)", "o": "Academic paper by Christian Szegedy, Wojciech Zaremba, Ilya Sutskever, Joan Bruna, Dumitru Erhan, Ian Goodfellow, and 1 others (from the arXiv API)", "dt": "2013-12-21 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1312.6199", "n": 184, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0183", "t": "Explaining and Harnessing Adversarial Examples (arXiv 1412.6572)", "o": "Academic paper by Ian J. Goodfellow, Jonathon Shlens, Christian Szegedy (from the arXiv API)", "dt": "2014-12-20 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1412.6572", "n": 185, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0184", "t": "DeepFool: a simple and accurate method to fool deep neural networks (arXiv 1511.04599)", "o": "Academic paper by Seyed-Mohsen Moosavi-Dezfooli, Alhussein Fawzi, Pascal Frossard (from the arXiv API)", "dt": "2015-11-14 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1511.04599", "n": 186, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0185", "t": "Transferability in Machine Learning: from Phenomena to Black-Box Attacks using Adversarial Samples (arXiv 1605.07277)", "o": "Academic paper by Nicolas Papernot, Patrick McDaniel, Ian Goodfellow (from the arXiv API)", "dt": "2016-05-24 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1605.07277", "n": 187, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0186", "t": "Curie: A method for protecting SVM Classifier from Poisoning Attack (arXiv 1606.01584)", "o": "Academic paper by Ricky Laishram, Vir Virander Phoha (from the arXiv API)", "dt": "2016-06-05 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1606.01584", "n": 188, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0187", "t": "Deep Learning with Differential Privacy (arXiv 1607.00133)", "o": "Academic paper by Martín Abadi, Andy Chu, Ian Goodfellow, H. Brendan McMahan, Ilya Mironov, Kunal Talwar, and 1 others (from the arXiv API)", "dt": "2016-07-01 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1607.00133", "n": 189, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0188", "t": "Adversarial examples in the physical world (arXiv 1607.02533)", "o": "Academic paper by Alexey Kurakin, Ian Goodfellow, Samy Bengio (from the arXiv API)", "dt": "2016-07-08 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1607.02533", "n": 190, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0189", "t": "Defensive Distillation is Not Robust to Adversarial Examples (arXiv 1607.04311)", "o": "Academic paper by Nicholas Carlini, David Wagner (from the arXiv API)", "dt": "2016-07-14 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1607.04311", "n": 191, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0190", "t": "Early Methods for Detecting Adversarial Images (arXiv 1608.00530)", "o": "Academic paper by Dan Hendrycks, Kevin Gimpel (from the arXiv API)", "dt": "2016-08-01 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1608.00530", "n": 192, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0191", "t": "Towards Evaluating the Robustness of Neural Networks (arXiv 1608.04644)", "o": "Academic paper by Nicholas Carlini, David Wagner (from the arXiv API)", "dt": "2016-08-16 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1608.04644", "n": 193, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0192", "t": "Stealing Machine Learning Models via Prediction APIs (arXiv 1609.02943)", "o": "Academic paper by Florian Tramèr, Fan Zhang, Ari Juels, Michael K. Reiter, Thomas Ristenpart (from the arXiv API)", "dt": "2016-09-09 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1609.02943", "n": 194, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0193", "t": "Membership Inference Attacks against Machine Learning Models (arXiv 1610.05820)", "o": "Academic paper by Reza Shokri, Marco Stronati, Congzheng Song, Vitaly Shmatikov (from the arXiv API)", "dt": "2016-10-18 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1610.05820", "n": 195, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0194", "t": "Universal adversarial perturbations (arXiv 1610.08401)", "o": "Academic paper by Seyed-Mohsen Moosavi-Dezfooli, Alhussein Fawzi, Omar Fawzi, Pascal Frossard (from the arXiv API)", "dt": "2016-10-26 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1610.08401", "n": 196, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0195", "t": "Delving into Transferable Adversarial Examples and Black-box Attacks (arXiv 1611.02770)", "o": "Academic paper by Yanpei Liu, Xinyun Chen, Chang Liu, Dawn Song (from the arXiv API)", "dt": "2016-11-08 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1611.02770", "n": 197, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0196", "t": "Reluplex: An Efficient SMT Solver for Verifying Deep Neural Networks (arXiv 1702.01135)", "o": "Academic paper by Guy Katz, Clark Barrett, David Dill, Kyle Julian, Mykel Kochenderfer (from the arXiv API)", "dt": "2017-02-03 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1702.01135", "n": 198, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0197", "t": "On Detecting Adversarial Perturbations (arXiv 1702.04267)", "o": "Academic paper by Jan Hendrik Metzen, Tim Genewein, Volker Fischer, Bastian Bischoff (from the arXiv API)", "dt": "2017-02-14 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1702.04267", "n": 199, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0198", "t": "Detecting Adversarial Samples from Artifacts (arXiv 1703.00410)", "o": "Academic paper by Reuben Feinman, Ryan R. Curtin, Saurabh Shintre, Andrew B. Gardner (from the arXiv API)", "dt": "2017-03-01 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1703.00410", "n": 200, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0199", "t": "The Space of Transferable Adversarial Examples (arXiv 1704.03453)", "o": "Academic paper by Florian Tramèr, Nicolas Papernot, Ian Goodfellow, Dan Boneh, Patrick McDaniel (from the arXiv API)", "dt": "2017-04-11 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1704.03453", "n": 201, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0200", "t": "Keeping the Bad Guys Out: Protecting and Vaccinating Deep Learning with JPEG Compression (arXiv 1705.02900)", "o": "Academic paper by Nilaksh Das, Madhuri Shanbhogue, Shang-Tse Chen, Fred Hohman, Li Chen, Michael E. Kounavis, and 1 others (from the arXiv API)", "dt": "2017-05-08 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1705.02900", "n": 202, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0201", "t": "Extending Defensive Distillation (arXiv 1705.05264)", "o": "Academic paper by Nicolas Papernot, Patrick McDaniel (from the arXiv API)", "dt": "2017-05-15 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1705.05264", "n": 203, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0202", "t": "Certified Defenses for Data Poisoning Attacks (arXiv 1706.03691)", "o": "Academic paper by Jacob Steinhardt, Pang Wei Koh, Percy Liang (from the arXiv API)", "dt": "2017-06-09 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1706.03691", "n": 204, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0203", "t": "Towards Deep Learning Models Resistant to Adversarial Attacks (arXiv 1706.06083)", "o": "Academic paper by Aleksander Madry, Aleksandar Makelov, Ludwig Schmidt, Dimitris Tsipras, Adrian Vladu (from the arXiv API)", "dt": "2017-06-19 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1706.06083", "n": 205, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0204", "t": "NO Need to Worry about Adversarial Examples in Object Detection in Autonomous Vehicles (arXiv 1707.03501)", "o": "Academic paper by Jiajun Lu, Hussein Sibai, Evan Fabry, David Forsyth (from the arXiv API)", "dt": "2017-07-12 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1707.03501", "n": 206, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0205", "t": "Synthesizing Robust Adversarial Examples (arXiv 1707.07397)", "o": "Academic paper by Anish Athalye, Logan Engstrom, Andrew Ilyas, Kevin Kwok (from the arXiv API)", "dt": "2017-07-24 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1707.07397", "n": 207, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0206", "t": "Robust Physical-World Attacks on Deep Learning Models (arXiv 1707.08945)", "o": "Academic paper by Kevin Eykholt, Ivan Evtimov, Earlence Fernandes, Bo Li, Amir Rahmati, Chaowei Xiao, and 3 others (from the arXiv API)", "dt": "2017-07-27 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1707.08945", "n": 208, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0207", "t": "BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chain (arXiv 1708.06733)", "o": "Academic paper by Tianyu Gu, Brendan Dolan-Gavitt, Siddharth Garg (from the arXiv API)", "dt": "2017-08-22 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1708.06733", "n": 209, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0208", "t": "Towards Poisoning of Deep Learning Algorithms with Back-gradient Optimization (arXiv 1708.08689)", "o": "Academic paper by Luis Muñoz-González, Battista Biggio, Ambra Demontis, Andrea Paudice, Vasin Wongrassamee, Emil C. Lupu, and 1 others (from the arXiv API)", "dt": "2017-08-29 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1708.08689", "n": 210, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0209", "t": "Privacy Risk in Machine Learning: Analyzing the Connection to Overfitting (arXiv 1709.01604)", "o": "Academic paper by Samuel Yeom, Irene Giacomelli, Matt Fredrikson, Somesh Jha (from the arXiv API)", "dt": "2017-09-05 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1709.01604", "n": 211, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0210", "t": "Fooling Vision and Language Models Despite Localization and Attention Mechanism (arXiv 1709.08693)", "o": "Academic paper by Xiaojun Xu, Xinyun Chen, Chang Liu, Anna Rohrbach, Trevor Darrell, Dawn Song (from the arXiv API)", "dt": "2017-09-25 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1709.08693", "n": 212, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0211", "t": "Mitigating Adversarial Effects Through Randomization (arXiv 1711.01991)", "o": "Academic paper by Cihang Xie, Jianyu Wang, Zhishuai Zhang, Zhou Ren, Alan Yuille (from the arXiv API)", "dt": "2017-11-06 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1711.01991", "n": 213, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0212", "t": "Attacking Visual Language Grounding with Adversarial Examples: A Case Study on Neural Image Captioning (arXiv 1712.02051)", "o": "Academic paper by Hongge Chen, Huan Zhang, Pin-Yu Chen, Jinfeng Yi, Cho-Jui Hsieh (from the arXiv API)", "dt": "2017-12-06 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1712.02051", "n": 214, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0213", "t": "Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoning (arXiv 1712.05526)", "o": "Academic paper by Xinyun Chen, Chang Liu, Bo Li, Kimberly Lu, Dawn Song (from the arXiv API)", "dt": "2017-12-15 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1712.05526", "n": 215, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0214", "t": "HotFlip: White-Box Adversarial Examples for Text Classification (arXiv 1712.06751)", "o": "Academic paper by Javid Ebrahimi, Anyi Rao, Daniel Lowd, Dejing Dou (from the arXiv API)", "dt": "2017-12-19 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1712.06751", "n": 216, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0215", "t": "Audio Adversarial Examples: Targeted Attacks on Speech-to-Text (arXiv 1801.01944)", "o": "Academic paper by Nicholas Carlini, David Wagner (from the arXiv API)", "dt": "2018-01-05 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1801.01944", "n": 217, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0216", "t": "Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examples (arXiv 1802.00420)", "o": "Academic paper by Anish Athalye, Nicholas Carlini, David Wagner (from the arXiv API)", "dt": "2018-02-01 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1802.00420", "n": 218, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0217", "t": "The Hidden Vulnerability of Distributed Learning in Byzantium (arXiv 1802.07927)", "o": "Academic paper by El Mahdi El Mhamdi, Rachid Guerraoui, Sébastien Rouault (from the arXiv API)", "dt": "2018-02-22 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1802.07927", "n": 219, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0218", "t": "The Secret Sharer: Evaluating and Testing Unintended Memorization in Neural Networks (arXiv 1802.08232)", "o": "Academic paper by Nicholas Carlini, Chang Liu, Úlfar Erlingsson, Jernej Kos, Dawn Song (from the arXiv API)", "dt": "2018-02-22 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1802.08232", "n": 220, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0219", "t": "Byzantine-Robust Distributed Learning: Towards Optimal Statistical Rates (arXiv 1803.01498)", "o": "Academic paper by Dong Yin, Yudong Chen, Kannan Ramchandran, Peter Bartlett (from the arXiv API)", "dt": "2018-03-05 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1803.01498", "n": 221, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0220", "t": "Sever: A Robust Meta-Algorithm for Stochastic Optimization (arXiv 1803.02815)", "o": "Academic paper by Ilias Diakonikolas, Gautam Kamath, Daniel M. Kane, Jerry Li, Jacob Steinhardt, Alistair Stewart (from the arXiv API)", "dt": "2018-03-07 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1803.02815", "n": 222, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0221", "t": "Deep k-Nearest Neighbors: Towards Confident, Interpretable and Robust Deep Learning (arXiv 1803.04765)", "o": "Academic paper by Nicolas Papernot, Patrick McDaniel (from the arXiv API)", "dt": "2018-03-13 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1803.04765", "n": 223, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0222", "t": "Technical Report: When Does Machine Learning FAIL? Generalized Transferability for Evasion and Poisoning Attacks (arXiv 1803.06975)", "o": "Academic paper by Octavian Suciu, Radu Mărginean, Yiğitcan Kaya, Hal Daumé, Tudor Dumitraş (from the arXiv API)", "dt": "2018-03-19 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1803.06975", "n": 224, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0223", "t": "Byzantine Stochastic Gradient Descent (arXiv 1803.08917)", "o": "Academic paper by Dan Alistarh, Zeyuan Allen-Zhu, Jerry Li (from the arXiv API)", "dt": "2018-03-23 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1803.08917", "n": 225, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0224", "t": "Manipulating Machine Learning: Poisoning Attacks and Countermeasures for Regression Learning (arXiv 1804.00308)", "o": "Academic paper by Matthew Jagielski, Alina Oprea, Battista Biggio, Chang Liu, Cristina Nita-Rotaru, Bo Li (from the arXiv API)", "dt": "2018-04-01 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1804.00308", "n": 226, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0225", "t": "Poison Frogs! Targeted Clean-Label Poisoning Attacks on Neural Networks (arXiv 1804.00792)", "o": "Academic paper by Ali Shafahi, W. Ronny Huang, Mahyar Najibi, Octavian Suciu, Christoph Studer, Tudor Dumitras, and 1 others (from the arXiv API)", "dt": "2018-04-03 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1804.00792", "n": 227, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0226", "t": "Is feature selection secure against training data poisoning? (arXiv 1804.07933)", "o": "Academic paper by Huang Xiao, Battista Biggio, Gavin Brown, Giorgio Fumera, Claudia Eckert, Fabio Roli (from the arXiv API)", "dt": "2018-04-21 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1804.07933", "n": 228, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0227", "t": "Black-box Adversarial Attacks with Limited Queries and Information (arXiv 1804.08598)", "o": "Academic paper by Andrew Ilyas, Logan Engstrom, Anish Athalye, Jessy Lin (from the arXiv API)", "dt": "2018-04-23 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1804.08598", "n": 229, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0228", "t": "Robustness May Be at Odds with Accuracy (arXiv 1805.12152)", "o": "Academic paper by Dimitris Tsipras, Shibani Santurkar, Logan Engstrom, Alexander Turner, Aleksander Madry (from the arXiv API)", "dt": "2018-05-30 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1805.12152", "n": 230, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0229", "t": "Adversarial Perturbations Against Real-Time Video Classification Systems (arXiv 1807.00458)", "o": "Academic paper by Shasha Li, Ajaya Neupane, Sujoy Paul, Chengyu Song, Srikanth V. Krishnamurthy, Amit K. Roy Chowdhury, and 1 others (from the arXiv API)", "dt": "2018-07-02 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1807.00458", "n": 231, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0230", "t": "Query-Efficient Hard-label Black-box Attack:An Optimization-based Approach (arXiv 1807.04457)", "o": "Academic paper by Minhao Cheng, Thong Le, Pin-Yu Chen, Jinfeng Yi, Huan Zhang, Cho-Jui Hsieh (from the arXiv API)", "dt": "2018-07-12 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1807.04457", "n": 232, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0231", "t": "Physical Adversarial Examples for Object Detectors (arXiv 1807.07769)", "o": "Academic paper by Kevin Eykholt, Ivan Evtimov, Earlence Fernandes, Bo Li, Amir Rahmati, Florian Tramer, and 3 others (from the arXiv API)", "dt": "2018-07-20 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1807.07769", "n": 233, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0232", "t": "Prior Convictions: Black-Box Adversarial Attacks with Bandits and Priors (arXiv 1807.07978)", "o": "Academic paper by Andrew Ilyas, Logan Engstrom, Aleksander Madry (from the arXiv API)", "dt": "2018-07-20 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1807.07978", "n": 234, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0233", "t": "Exploring Adversarial Examples in Malware Detection (arXiv 1810.08280)", "o": "Academic paper by Octavian Suciu, Scott E. Coull, Jeffrey Johns (from the arXiv API)", "dt": "2018-10-18 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1810.08280", "n": 235, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0234", "t": "A Marauder's Map of Security and Privacy in Machine Learning (arXiv 1811.01134)", "o": "Academic paper by Nicolas Papernot (from the arXiv API)", "dt": "2018-11-03 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1811.01134", "n": 236, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0235", "t": "Exploring Connections Between Active Learning and Model Extraction (arXiv 1811.02054)", "o": "Academic paper by Varun Chandrasekaran, Kamalika Chaudhuri, Irene Giacomelli, Somesh Jha, Songbai Yan (from the arXiv API)", "dt": "2018-11-05 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1811.02054", "n": 237, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0236", "t": "Detecting Backdoor Attacks on Deep Neural Networks by Activation Clustering (arXiv 1811.03728)", "o": "Academic paper by Bryant Chen, Wilka Carvalho, Nathalie Baracaldo, Heiko Ludwig, Benjamin Edwards, Taesung Lee, and 2 others (from the arXiv API)", "dt": "2018-11-09 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1811.03728", "n": 238, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0237", "t": "Analyzing Federated Learning through an Adversarial Lens (arXiv 1811.12470)", "o": "Academic paper by Arjun Nitin Bhagoji, Supriyo Chakraborty, Prateek Mittal, Seraphin Calo (from the arXiv API)", "dt": "2018-11-29 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1811.12470", "n": 239, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0238", "t": "Transferable Adversarial Attacks for Image and Video Object Detection (arXiv 1811.12641)", "o": "Academic paper by Xingxing Wei, Siyuan Liang, Ning Chen, Xiaochun Cao (from the arXiv API)", "dt": "2018-11-30 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1811.12641", "n": 240, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0239", "t": "SentiNet: Detecting Localized Universal Attacks Against Deep Learning Systems (arXiv 1812.00292)", "o": "Academic paper by Edward Chou, Florian Tramèr, Giancarlo Pellegrino (from the arXiv API)", "dt": "2018-12-02 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1812.00292", "n": 241, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0240", "t": "Comprehensive Privacy Analysis of Deep Learning: Passive and Active White-box Inference Attacks against Centralized and Federated Learning (arXiv 1812.00910)", "o": "Academic paper by Milad Nasr, Reza Shokri, Amir Houmansadr (from the arXiv API)", "dt": "2018-12-03 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1812.00910", "n": 242, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0241", "t": "Theoretically Principled Trade-off between Robustness and Accuracy (arXiv 1901.08573)", "o": "Academic paper by Hongyang Zhang, Yaodong Yu, Jiantao Jiao, Eric P. Xing, Laurent El Ghaoui, Michael I. Jordan (from the arXiv API)", "dt": "2019-01-24 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1901.08573", "n": 243, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0242", "t": "Certified Adversarial Robustness via Randomized Smoothing (arXiv 1902.02918)", "o": "Academic paper by Jeremy M Cohen, Elan Rosenfeld, J. Zico Kolter (from the arXiv API)", "dt": "2019-02-08 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1902.02918", "n": 244, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0243", "t": "STRIP: A Defence Against Trojan Attacks on Deep Neural Networks (arXiv 1902.06531)", "o": "Academic paper by Yansong Gao, Chang Xu, Derui Wang, Shiping Chen, Damith C. Ranasinghe, Surya Nepal (from the arXiv API)", "dt": "2019-02-18 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1902.06531", "n": 245, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0244", "t": "Evaluating Differentially Private Machine Learning in Practice (arXiv 1902.08874)", "o": "Academic paper by Bargav Jayaraman, David Evans (from the arXiv API)", "dt": "2019-02-24 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1902.08874", "n": 246, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0245", "t": "Rallying Adversarial Techniques against Deep Learning for Network Security (arXiv 1903.11688)", "o": "Academic paper by Joseph Clements, Yuzhe Yang, Ankur Sharma, Hongxin Hu, Yingjie Lao (from the arXiv API)", "dt": "2019-03-27 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1903.11688", "n": 247, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0246", "t": "Defensive Quantization: When Efficiency Meets Robustness (arXiv 1904.08444)", "o": "Academic paper by Ji Lin, Chuang Gan, Song Han (from the arXiv API)", "dt": "2019-04-17 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1904.08444", "n": 248, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0247", "t": "Adversarial Examples Are Not Bugs, They Are Features (arXiv 1905.02175)", "o": "Academic paper by Andrew Ilyas, Shibani Santurkar, Dimitris Tsipras, Logan Engstrom, Brandon Tran, Aleksander Madry (from the arXiv API)", "dt": "2019-05-06 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1905.02175", "n": 249, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0248", "t": "Transferable Clean-Label Poisoning Attacks on Deep Neural Nets (arXiv 1905.05897)", "o": "Academic paper by Chen Zhu, W. Ronny Huang, Ali Shafahi, Hengduo Li, Gavin Taylor, Christoph Studer, and 1 others (from the arXiv API)", "dt": "2019-05-15 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1905.05897", "n": 250, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0249", "t": "Parsimonious Black-Box Adversarial Attacks via Efficient Combinatorial Optimization (arXiv 1905.06635)", "o": "Academic paper by Seungyong Moon, Gaon An, Hyun Oh Song (from the arXiv API)", "dt": "2019-05-16 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1905.06635", "n": 251, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0250", "t": "Learning to Confuse: Generating Training Time Adversarial Data with Auto-Encoder (arXiv 1905.09027)", "o": "Academic paper by Ji Feng, Qi-Zhi Cai, Zhi-Hua Zhou (from the arXiv API)", "dt": "2019-05-22 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1905.09027", "n": 252, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0251", "t": "Stolen Memories: Leveraging Model Memorization for Calibrated White-Box Membership Inference (arXiv 1906.11798)", "o": "Academic paper by Klas Leino, Matt Fredrikson (from the arXiv API)", "dt": "2019-06-27 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1906.11798", "n": 253, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0252", "t": "Adversarial Objects Against LiDAR-Based Autonomous Driving Systems (arXiv 1907.05418)", "o": "Academic paper by Yulong Cao, Chaowei Xiao, Dawei Yang, Jing Fang, Ruigang Yang, Mingyan Liu, and 1 others (from the arXiv API)", "dt": "2019-07-11 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1907.05418", "n": 254, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0253", "t": "On Defending Against Label Flipping Attacks on Malware Detection Systems (arXiv 1908.04473)", "o": "Academic paper by Rahim Taheri, Reza Javidan, Mohammad Shojafar, Zahra Pooranian, Ali Miri, Mauro Conti (from the arXiv API)", "dt": "2019-08-13 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1908.04473", "n": 255, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0254", "t": "Universal Adversarial Triggers for Attacking and Analyzing NLP (arXiv 1908.07125)", "o": "Academic paper by Eric Wallace, Shi Feng, Nikhil Kandpal, Matt Gardner, Sameer Singh (from the arXiv API)", "dt": "2019-08-20 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1908.07125", "n": 256, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0255", "t": "White-box vs Black-box: Bayes Optimal Strategies for Membership Inference (arXiv 1908.11229)", "o": "Academic paper by Alexandre Sablayrolles, Matthijs Douze, Yann Ollivier, Cordelia Schmid, Hervé Jégou (from the arXiv API)", "dt": "2019-08-29 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1908.11229", "n": 257, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0256", "t": "High Accuracy and High Fidelity Extraction of Neural Networks (arXiv 1909.01838)", "o": "Academic paper by Matthew Jagielski, Nicholas Carlini, David Berthelot, Alex Kurakin, Nicolas Papernot (from the arXiv API)", "dt": "2019-09-03 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1909.01838", "n": 258, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0257", "t": "Invisible Backdoor Attacks on Deep Neural Networks via Steganography and Regularization (arXiv 1909.02742)", "o": "Academic paper by Shaofeng Li, Minhui Xue, Benjamin Zi Hao Zhao, Haojin Zhu, Xinpeng Zhang (from the arXiv API)", "dt": "2019-09-06 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1909.02742", "n": 259, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0258", "t": "FENCE: Feasible Evasion Attacks on Neural Networks in Constrained Environments (arXiv 1909.10480)", "o": "Academic paper by Alesia Chernikova, Alina Oprea (from the arXiv API)", "dt": "2019-09-23 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1909.10480", "n": 260, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0259", "t": "Sign-OPT: A Query-Efficient Hard-label Adversarial Attack (arXiv 1909.10773)", "o": "Academic paper by Minhao Cheng, Simranjit Singh, Patrick Chen, Pin-Yu Chen, Sijia Liu, Cho-Jui Hsieh (from the arXiv API)", "dt": "2019-09-24 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1909.10773", "n": 261, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0260", "t": "Deep k-NN Defense against Clean-label Data Poisoning Attacks (arXiv 1909.13374)", "o": "Academic paper by Neehar Peri, Neal Gupta, W. Ronny Huang, Liam Fowl, Chen Zhu, Soheil Feizi, and 2 others (from the arXiv API)", "dt": "2019-09-29 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1909.13374", "n": 262, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0261", "t": "NeuronInspect: Detecting Backdoors in Neural Networks via Output Explanations (arXiv 1911.07399)", "o": "Academic paper by Xijie Huang, Moustafa Alzantot, Mani Srivastava (from the arXiv API)", "dt": "2019-11-18 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1911.07399", "n": 263, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0262", "t": "Can You Really Backdoor Federated Learning? (arXiv 1911.07963)", "o": "Academic paper by Ziteng Sun, Peter Kairouz, Ananda Theertha Suresh, H. Brendan McMahan (from the arXiv API)", "dt": "2019-11-18 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1911.07963", "n": 264, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0263", "t": "Heuristic Black-box Adversarial Attacks on Video Recognition Models (arXiv 1911.09449)", "o": "Academic paper by Zhipeng Wei, Jingjing Chen, Xingxing Wei, Linxi Jiang, Tat-Seng Chua, Fengfeng Zhou, and 1 others (from the arXiv API)", "dt": "2019-11-21 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1911.09449", "n": 265, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0264", "t": "Failure Modes in Machine Learning Systems (arXiv 1911.11034)", "o": "Academic paper by Ram Shankar Siva Kumar, David O Brien, Kendra Albert, Salomé Viljöen, Jeffrey Snover (from the arXiv API)", "dt": "2019-11-25 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/1911.11034", "n": 266, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0265", "t": "Fundamental Tradeoffs between Invariance and Sensitivity to Adversarial Perturbations (arXiv 2002.04599)", "o": "Academic paper by Florian Tramèr, Jens Behrmann, Nicholas Carlini, Nicolas Papernot, Jörn-Henrik Jacobsen (from the arXiv API)", "dt": "2020-02-11 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2002.04599", "n": 267, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0266", "t": "Fast Geometric Projections for Local Robustness Certification (arXiv 2002.04742)", "o": "Academic paper by Aymeric Fromherz, Klas Leino, Matt Fredrikson, Bryan Parno, Corina Păsăreanu (from the arXiv API)", "dt": "2020-02-12 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2002.04742", "n": 268, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0267", "t": "Adversarial Machine Learning -- Industry Perspectives (arXiv 2002.05646)", "o": "Academic paper by Ram Shankar Siva Kumar, Magnus Nyström, John Lambert, Andrew Marshall, Mario Goertzel, Andi Comissoneru, and 2 others (from the arXiv API)", "dt": "2020-02-04 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2002.05646", "n": 269, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0268", "t": "On Adaptive Attacks to Adversarial Example Defenses (arXiv 2002.08347)", "o": "Academic paper by Florian Tramer, Nicholas Carlini, Wieland Brendel, Aleksander Madry (from the arXiv API)", "dt": "2020-02-19 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2002.08347", "n": 270, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0269", "t": "Distributed Momentum for Byzantine-resilient Learning (arXiv 2003.00010)", "o": "Academic paper by El-Mahdi El-Mhamdi, Rachid Guerraoui, Sébastien Rouault (from the arXiv API)", "dt": "2020-02-28 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2003.00010", "n": 271, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0270", "t": "Towards Backdoor Attacks and Defense in Robust Machine Learning Models (arXiv 2003.00865)", "o": "Academic paper by Ezekiel Soremekun, Sakshi Udeshi, Sudipta Chattopadhyay (from the arXiv API)", "dt": "2020-02-25 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2003.00865", "n": 272, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0271", "t": "Dynamic Backdoor Attacks Against Machine Learning Models (arXiv 2003.03675)", "o": "Academic paper by Ahmed Salem, Rui Wen, Michael Backes, Shiqing Ma, Yang Zhang (from the arXiv API)", "dt": "2020-03-07 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2003.03675", "n": 273, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0272", "t": "Cryptanalytic Extraction of Neural Network Models (arXiv 2003.04884)", "o": "Academic paper by Nicholas Carlini, Matthew Jagielski, Ilya Mironov (from the arXiv API)", "dt": "2020-03-10 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2003.04884", "n": 274, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0273", "t": "Breaking certified defenses: Semantic adversarial examples with spoofed robustness certificates (arXiv 2003.08937)", "o": "Academic paper by Amin Ghiasi, Ali Shafahi, Tom Goldstein (from the arXiv API)", "dt": "2020-03-19 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2003.08937", "n": 275, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0274", "t": "Information Leakage in Embedding Models (arXiv 2004.00053)", "o": "Academic paper by Congzheng Song, Ananth Raghunathan (from the arXiv API)", "dt": "2020-03-31 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2004.00053", "n": 276, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0275", "t": "MetaPoison: Practical General-purpose Clean-label Data Poisoning (arXiv 2004.00225)", "o": "Academic paper by W. Ronny Huang, Jonas Geiping, Liam Fowl, Gavin Taylor, Tom Goldstein (from the arXiv API)", "dt": "2020-04-01 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2004.00225", "n": 277, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0276", "t": "Weight Poisoning Attacks on Pre-trained Models (arXiv 2004.06660)", "o": "Academic paper by Keita Kurita, Paul Michel, Graham Neubig (from the arXiv API)", "dt": "2020-04-14 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2004.06660", "n": 278, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0277", "t": "Investigating Vulnerability to Adversarial Examples on Multimodal Data Fusion in Deep Learning (arXiv 2005.10987)", "o": "Academic paper by Youngjoon Yu, Hong Joo Lee, Byeong Cheon Kim, Jung Uk Kim, Yong Man Ro (from the arXiv API)", "dt": "2020-05-22 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2005.10987", "n": 279, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0278", "t": "Sponge Examples: Energy-Latency Attacks on Neural Networks (arXiv 2006.03463)", "o": "Academic paper by Ilia Shumailov, Yiren Zhao, Daniel Bates, Nicolas Papernot, Robert Mullins, Ross Anderson (from the arXiv API)", "dt": "2020-06-05 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2006.03463", "n": 280, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0279", "t": "Leakage of Dataset Properties in Multi-Party Machine Learning (arXiv 2006.07267)", "o": "Academic paper by Wanrong Zhang, Shruti Tople, Olga Ohrimenko (from the arXiv API)", "dt": "2020-06-12 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2006.07267", "n": 281, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0280", "t": "Auditing Differentially Private Machine Learning: How Private is Private SGD? (arXiv 2006.07709)", "o": "Academic paper by Matthew Jagielski, Jonathan Ullman, Alina Oprea (from the arXiv API)", "dt": "2020-06-13 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2006.07709", "n": 282, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0281", "t": "Just How Toxic is Data Poisoning? A Unified Benchmark for Backdoor and Data Poisoning Attacks (arXiv 2006.12557)", "o": "Academic paper by Avi Schwarzschild, Micah Goldblum, Arjun Gupta, John P Dickerson, Tom Goldstein (from the arXiv API)", "dt": "2020-06-22 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2006.12557", "n": 283, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0282", "t": "Subpopulation Data Poisoning Attacks (arXiv 2006.14026)", "o": "Academic paper by Matthew Jagielski, Giorgio Severi, Niklas Pousette Harger, Alina Oprea (from the arXiv API)", "dt": "2020-06-24 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2006.14026", "n": 284, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0283", "t": "Backdoor Attacks Against Deep Learning Systems in the Physical World (arXiv 2006.14580)", "o": "Academic paper by Emily Wenger, Josephine Passananti, Arjun Bhagoji, Yuanshun Yao, Haitao Zheng, Ben Y. Zhao (from the arXiv API)", "dt": "2020-06-25 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2006.14580", "n": 285, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0284", "t": "Deep Partition Aggregation: Provable Defense against General Poisoning Attacks (arXiv 2006.14768)", "o": "Academic paper by Alexander Levine, Soheil Feizi (from the arXiv API)", "dt": "2020-06-26 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2006.14768", "n": 286, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0285", "t": "Reflection Backdoor: A Natural Backdoor Attack on Deep Neural Networks (arXiv 2007.02343)", "o": "Academic paper by Yunfei Liu, Xingjun Ma, James Bailey, Feng Lu (from the arXiv API)", "dt": "2020-07-05 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2007.02343", "n": 287, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0286", "t": "Attack of the Tails: Yes, You Really Can Backdoor Federated Learning (arXiv 2007.05084)", "o": "Academic paper by Hongyi Wang, Kartik Sreenivasan, Shashank Rajput, Harit Vishwakarma, Saurabh Agarwal, Jy-yong Sohn, and 2 others (from the arXiv API)", "dt": "2020-07-09 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2007.05084", "n": 288, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0287", "t": "ML Privacy Meter: Aiding Regulatory Compliance by Quantifying the Privacy Risks of Machine Learning (arXiv 2007.09339)", "o": "Academic paper by Sasi Kumar Murakonda, Reza Shokri (from the arXiv API)", "dt": "2020-07-18 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2007.09339", "n": 289, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0288", "t": "Label-Only Membership Inference Attacks (arXiv 2007.14321)", "o": "Academic paper by Christopher A. Choquette-Choo, Florian Tramer, Nicholas Carlini, Nicolas Papernot (from the arXiv API)", "dt": "2020-07-28 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2007.14321", "n": 290, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0289", "t": "Collaborative Learning in the Jungle (Decentralized, Byzantine, Heterogeneous, Asynchronous and Nonconvex Learning) (arXiv 2008.00742)", "o": "Academic paper by El-Mahdi El-Mhamdi, Sadegh Farhadkhani, Rachid Guerraoui, Arsany Guirguis, Lê Nguyên Hoang, Sébastien Rouault (from the arXiv API)", "dt": "2020-08-03 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2008.00742", "n": 291, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0290", "t": "SoK: Certified Robustness for Deep Neural Networks (arXiv 2009.04131)", "o": "Academic paper by Linyi Li, Tao Xie, Bo Li (from the arXiv API)", "dt": "2020-09-09 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2009.04131", "n": 292, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0291", "t": "Garfield: System Support for Byzantine Machine Learning (arXiv 2010.05888)", "o": "Academic paper by Rachid Guerraoui, Arsany Guirguis, Jérémy Max Plassmann, Anton Alexandre Ragot, Sébastien Rouault (from the arXiv API)", "dt": "2020-10-12 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2010.05888", "n": 293, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0292", "t": "RobustBench: a standardized adversarial robustness benchmark (arXiv 2010.09670)", "o": "Academic paper by Francesco Croce, Maksym Andriushchenko, Vikash Sehwag, Edoardo Debenedetti, Nicolas Flammarion, Mung Chiang, and 2 others (from the arXiv API)", "dt": "2020-10-19 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2010.09670", "n": 294, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0293", "t": "Once-for-All Adversarial Training: In-Situ Tradeoff between Robustness and Accuracy for Free (arXiv 2010.11828)", "o": "Academic paper by Haotao Wang, Tianlong Chen, Shupeng Gui, Ting-Kuei Hu, Ji Liu, Zhangyang Wang (from the arXiv API)", "dt": "2020-10-22 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2010.11828", "n": 295, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0294", "t": "Concealed Data Poisoning Attacks on NLP Models (arXiv 2010.12563)", "o": "Academic paper by Eric Wallace, Tony Z. Zhao, Shi Feng, Sameer Singh (from the arXiv API)", "dt": "2020-10-23 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2010.12563", "n": 296, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0295", "t": "Towards Defending Multiple $\\ell_p$-norm Bounded Adversarial Perturbations via Gated Batch Normalization (arXiv 2012.01654)", "o": "Academic paper by Aishan Liu, Shiyu Tang, Xinyun Chen, Lei Huang, Haotong Qin, Xianglong Liu, and 1 others (from the arXiv API)", "dt": "2020-12-03 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2012.01654", "n": 297, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0296", "t": "FLTrust: Byzantine-robust Federated Learning via Trust Bootstrapping (arXiv 2012.13995)", "o": "Academic paper by Xiaoyu Cao, Minghong Fang, Jia Liu, Neil Zhenqiang Gong (from the arXiv API)", "dt": "2020-12-27 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2012.13995", "n": 298, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0297", "t": "Adversary Instantiation: Lower Bounds for Differentially Private Machine Learning (arXiv 2101.04535)", "o": "Academic paper by Milad Nasr, Shuang Song, Abhradeep Thakurta, Nicolas Papernot, Nicholas Carlini (from the arXiv API)", "dt": "2021-01-11 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2101.04535", "n": 299, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0298", "t": "Preventing Unauthorized Use of Proprietary Data: Poisoning for Secure Dataset Release (arXiv 2103.02683)", "o": "Academic paper by Liam Fowl, Ping-yeh Chiang, Micah Goldblum, Jonas Geiping, Arpit Bansal, Wojtek Czaja, and 1 others (from the arXiv API)", "dt": "2021-02-16 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2103.02683", "n": 300, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0299", "t": "SPECTRE: Defending Against Backdoor Attacks Using Robust Statistics (arXiv 2104.11315)", "o": "Academic paper by Jonathan Hayase, Weihao Kong, Raghav Somani, Sewoong Oh (from the arXiv API)", "dt": "2021-04-22 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2104.11315", "n": 301, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0300", "t": "We Can Always Catch You: Detecting Adversarial Patched Objects WITH or WITHOUT Signature (arXiv 2106.05261)", "o": "Academic paper by Jiachun Li, Jianan Feng, Jianjun Huang, Bin Liang (from the arXiv API)", "dt": "2021-06-09 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2106.05261", "n": 302, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0301", "t": "Asleep at the Keyboard? Assessing the Security of GitHub Copilot's Code Contributions (arXiv 2108.09293)", "o": "Academic paper by Hammond Pearce, Baleegh Ahmad, Benjamin Tan, Brendan Dolan-Gavitt, Ramesh Karri (from the arXiv API)", "dt": "2021-08-20 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2108.09293", "n": 303, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0302", "t": "Formalizing and Estimating Distribution Inference Risks (arXiv 2109.06024)", "o": "Academic paper by Anshuman Suri, David Evans (from the arXiv API)", "dt": "2021-09-13 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2109.06024", "n": 304, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0303", "t": "Adversarial Unlearning of Backdoors via Implicit Hypergradient (arXiv 2110.03735)", "o": "Academic paper by Yi Zeng, Si Chen, Won Park, Z. Morley Mao, Ming Jin, Ruoxi Jia (from the arXiv API)", "dt": "2021-10-07 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2110.03735", "n": 305, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0304", "t": "Trigger Hunting with a Topological Prior for Trojan Detection (arXiv 2110.08335)", "o": "Academic paper by Xiaoling Hu, Xiao Lin, Michael Cogswell, Yi Yao, Susmit Jha, Chao Chen (from the arXiv API)", "dt": "2021-10-15 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2110.08335", "n": 306, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0305", "t": "FL-WBC: Enhancing Robustness against Model Poisoning Attacks in Federated Learning from a Client Perspective (arXiv 2110.13864)", "o": "Academic paper by Jingwei Sun, Ang Li, Louis DiValentin, Amin Hassanzadeh, Yiran Chen, Hai Li (from the arXiv API)", "dt": "2021-10-26 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2110.13864", "n": 307, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0306", "t": "Adversarial Neuron Pruning Purifies Backdoored Deep Models (arXiv 2110.14430)", "o": "Academic paper by Dongxian Wu, Yisen Wang (from the arXiv API)", "dt": "2021-10-27 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2110.14430", "n": 308, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0307", "t": "Property Inference Attacks Against GANs (arXiv 2111.07608)", "o": "Academic paper by Junhao Zhou, Yufei Chen, Chao Shen, Yang Zhang (from the arXiv API)", "dt": "2021-11-15 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2111.07608", "n": 309, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0308", "t": "When the Curious Abandon Honesty: Federated Learning Is Not Private (arXiv 2112.02918)", "o": "Academic paper by Franziska Boenisch, Adam Dziedzic, Roei Schuster, Ali Shahin Shamsabadi, Ilia Shumailov, Nicolas Papernot (from the arXiv API)", "dt": "2021-12-06 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2112.02918", "n": 310, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0309", "t": "The King is Naked: on the Notion of Robustness for Natural Language Processing (arXiv 2112.07605)", "o": "Academic paper by Emanuele La Malfa, Marta Kwiatkowska (from the arXiv API)", "dt": "2021-12-13 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2112.07605", "n": 311, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0310", "t": "Reconstructing Training Data with Informed Adversaries (arXiv 2201.04845)", "o": "Academic paper by Borja Balle, Giovanni Cherubin, Jamie Hayes (from the arXiv API)", "dt": "2022-01-13 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2201.04845", "n": 312, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0311", "t": "Post-Training Detection of Backdoor Attacks for Two-Class and Multi-Attack Scenarios (arXiv 2201.08474)", "o": "Academic paper by Zhen Xiang, David J. Miller, George Kesidis (from the arXiv API)", "dt": "2022-01-20 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2201.08474", "n": 313, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0312", "t": "Improved Certified Defenses against Data Poisoning with (Deterministic) Finite Aggregation (arXiv 2202.02628)", "o": "Academic paper by Wenxiao Wang, Alexander Levine, Soheil Feizi (from the arXiv API)", "dt": "2022-02-05 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2202.02628", "n": 314, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0313", "t": "Red Teaming Language Models with Language Models (arXiv 2202.03286)", "o": "Academic paper by Ethan Perez, Saffron Huang, Francis Song, Trevor Cai, Roman Ring, John Aslanides, and 3 others (from the arXiv API)", "dt": "2022-02-07 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2202.03286", "n": 315, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0314", "t": "Jigsaw Puzzle: Selective Backdoor Attack to Subvert Malware Classifiers (arXiv 2202.05470)", "o": "Academic paper by Limin Yang, Zhi Chen, Jacopo Cortellazzi, Feargus Pendlebury, Kevin Tu, Fabio Pierazzi, and 2 others (from the arXiv API)", "dt": "2022-02-11 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2202.05470", "n": 316, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0315", "t": "Quantifying Memorization Across Neural Language Models (arXiv 2202.07646)", "o": "Academic paper by Nicholas Carlini, Daphne Ippolito, Matthew Jagielski, Katherine Lee, Florian Tramer, Chiyuan Zhang (from the arXiv API)", "dt": "2022-02-15 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2202.07646", "n": 317, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0316", "t": "Bounding Membership Inference (arXiv 2202.12232)", "o": "Academic paper by Anvith Thudi, Ilia Shumailov, Franziska Boenisch, Nicolas Papernot (from the arXiv API)", "dt": "2022-02-24 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2202.12232", "n": 318, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0317", "t": "Planting Undetectable Backdoors in Machine Learning Models (arXiv 2204.06974)", "o": "Academic paper by Shafi Goldwasser, Michael P. Kim, Vinod Vaikuntanathan, Or Zamir (from the arXiv API)", "dt": "2022-04-14 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2204.06974", "n": 319, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0318", "t": "Just Fine-tune Twice: Selective Differential Privacy for Large Language Models (arXiv 2204.07667)", "o": "Academic paper by Weiyan Shi, Ryan Shea, Si Chen, Chiyuan Zhang, Ruoxi Jia, Zhou Yu (from the arXiv API)", "dt": "2022-04-15 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2204.07667", "n": 320, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0319", "t": "Indiscriminate Data Poisoning Attacks on Neural Networks (arXiv 2204.09092)", "o": "Academic paper by Yiwei Lu, Gautam Kamath, Yaoliang Yu (from the arXiv API)", "dt": "2022-04-19 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2204.09092", "n": 321, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0320", "t": "Are Large Pre-Trained Language Models Leaking Your Personal Information? (arXiv 2205.12628)", "o": "Academic paper by Jie Huang, Hanyin Shao, Kevin Chen-Chuan Chang (from the arXiv API)", "dt": "2022-05-25 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2205.12628", "n": 322, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0321", "t": "Support Vector Machines under Adversarial Label Contamination (arXiv 2206.00352)", "o": "Academic paper by Huang Xiao, Battista Biggio, Blaine Nelson, Han Xiao, Claudia Eckert, Fabio Roli (from the arXiv API)", "dt": "2022-06-01 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2206.00352", "n": 323, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0322", "t": "Reconstructing Training Data from Trained Neural Networks (arXiv 2206.07758)", "o": "Academic paper by Niv Haim, Gal Vardi, Gilad Yehudai, Ohad Shamir, Michal Irani (from the arXiv API)", "dt": "2022-06-15 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2206.07758", "n": 324, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0323", "t": "Neurotoxin: Durable Backdoors in Federated Learning (arXiv 2206.10341)", "o": "Academic paper by Zhengming Zhang, Ashwinee Panda, Linyue Song, Yaoqing Yang, Michael W. Mahoney, Joseph E. Gonzalez, and 2 others (from the arXiv API)", "dt": "2022-06-12 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2206.10341", "n": 325, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0324", "t": "(Certified!!) Adversarial Robustness for Free! (arXiv 2206.10550)", "o": "Academic paper by Nicholas Carlini, Florian Tramer, Krishnamurthy Dj Dvijotham, Leslie Rice, Mingjie Sun, J. Zico Kolter (from the arXiv API)", "dt": "2022-06-21 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2206.10550", "n": 326, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0325", "t": "Defending Multimodal Fusion Models against Single-Source Adversaries (arXiv 2206.12714)", "o": "Academic paper by Karren Yang, Wan-Yi Lin, Manash Barman, Filipe Condessa, Zico Kolter (from the arXiv API)", "dt": "2022-06-25 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2206.12714", "n": 327, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0326", "t": "Threat Assessment in Machine Learning based Systems (arXiv 2207.00091)", "o": "Academic paper by Lionel Nganyewou Tidjon, Foutse Khomh (from the arXiv API)", "dt": "2022-06-30 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2207.00091", "n": 328, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0327", "t": "SNAP: Efficient Extraction of Private Properties with Poisoning (arXiv 2208.12348)", "o": "Academic paper by Harsh Chaudhari, John Abascal, Alina Oprea, Matthew Jagielski, Florian Tramèr, Jonathan Ullman (from the arXiv API)", "dt": "2022-08-25 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2208.12348", "n": 329, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0328", "t": "Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons Learned (arXiv 2209.07858)", "o": "Academic paper by Deep Ganguli, Liane Lovitt, Jackson Kernion, Amanda Askell, Yuntao Bai, Saurav Kadavath, and 30 others (from the arXiv API)", "dt": "2022-08-23 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2209.07858", "n": 330, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0329", "t": "Text Revealer: Private Text Reconstruction via Model Inversion Attacks against Transformers (arXiv 2209.10505)", "o": "Academic paper by Ruisi Zhang, Seira Hidano, Farinaz Koushanfar (from the arXiv API)", "dt": "2022-09-21 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2209.10505", "n": 331, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0330", "t": "Do Users Write More Insecure Code with AI Assistants? (arXiv 2211.03622)", "o": "Academic paper by Neil Perry, Megha Srivastava, Deepak Kumar, Dan Boneh (from the arXiv API)", "dt": "2022-11-07 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2211.03622", "n": 332, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0331", "t": "How to Backdoor Diffusion Models? (arXiv 2212.05400)", "o": "Academic paper by Sheng-Yen Chou, Pin-Yu Chen, Tsung-Yi Ho (from the arXiv API)", "dt": "2022-12-11 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2212.05400", "n": 333, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0332", "t": "SoK: Let the Privacy Games Begin! A Unified Treatment of Data Inference Privacy in Machine Learning (arXiv 2212.10986)", "o": "Academic paper by Ahmed Salem, Giovanni Cherubin, David Evans, Boris Köpf, Andrew Paverd, Anshuman Suri, and 2 others (from the arXiv API)", "dt": "2022-12-21 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2212.10986", "n": 334, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0333", "t": "\"Real Attackers Don't Compute Gradients\": Bridging the Gap Between Adversarial ML Research and Practice (arXiv 2212.14315)", "o": "Academic paper by Giovanni Apruzzese, Hyrum S. Anderson, Savino Dambra, David Freeman, Fabio Pierazzi, Kevin A. Roundy (from the arXiv API)", "dt": "2022-12-29 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2212.14315", "n": 335, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0334", "t": "TrojanPuzzle: Covertly Poisoning Code-Suggestion Models (arXiv 2301.02344)", "o": "Academic paper by Hojjat Aghakhani, Wei Dai, Andre Manoel, Xavier Fernandes, Anant Kharkar, Christopher Kruegel, and 4 others (from the arXiv API)", "dt": "2023-01-06 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2301.02344", "n": 336, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0335", "t": "A Watermark for Large Language Models (arXiv 2301.10226)", "o": "Academic paper by John Kirchenbauer, Jonas Geiping, Yuxin Wen, Jonathan Katz, Ian Miers, Tom Goldstein (from the arXiv API)", "dt": "2023-01-24 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2301.10226", "n": 337, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0336", "t": "Extracting Training Data from Diffusion Models (arXiv 2301.13188)", "o": "Academic paper by Nicholas Carlini, Jamie Hayes, Milad Nasr, Matthew Jagielski, Vikash Sehwag, Florian Tramèr, and 3 others (from the arXiv API)", "dt": "2023-01-30 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2301.13188", "n": 338, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0337", "t": "Analyzing Leakage of Personally Identifiable Information in Language Models (arXiv 2302.00539)", "o": "Academic paper by Nils Lukas, Ahmed Salem, Robert Sim, Shruti Tople, Lukas Wutschitz, Santiago Zanella-Béguelin (from the arXiv API)", "dt": "2023-02-01 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2302.00539", "n": 339, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0338", "t": "One-shot Empirical Privacy Estimation for Federated Learning (arXiv 2302.03098)", "o": "Academic paper by Galen Andrew, Peter Kairouz, Sewoong Oh, Alina Oprea, H. Brendan McMahan, Vinith M. Suriyakumar (from the arXiv API)", "dt": "2023-02-06 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2302.03098", "n": 340, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0339", "t": "Prompt Stealing Attacks Against Text-to-Image Generation Models (arXiv 2302.09923)", "o": "Academic paper by Xinyue Shen, Yiting Qu, Michael Backes, Yang Zhang (from the arXiv API)", "dt": "2023-02-20 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2302.09923", "n": 341, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0340", "t": "Poisoning Web-Scale Training Datasets is Practical (arXiv 2302.10149)", "o": "Academic paper by Nicholas Carlini, Matthew Jagielski, Christopher A. Choquette-Choo, Daniel Paleka, Will Pearce, Hyrum Anderson, and 3 others (from the arXiv API)", "dt": "2023-02-20 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2302.10149", "n": 342, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0341", "t": "Red Teaming Deep Neural Networks with Feature Synthesis Tools (arXiv 2302.10894)", "o": "Academic paper by Stephen Casper, Yuxiao Li, Jiawei Li, Tong Bu, Kevin Zhang, Kaivalya Hariharan, and 1 others (from the arXiv API)", "dt": "2023-02-08 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2302.10894", "n": 343, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0342", "t": "Not what you've signed up for: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injection (arXiv 2302.12173)", "o": "Academic paper by Kai Greshake, Sahar Abdelnabi, Shailesh Mishra, Christoph Endres, Thorsten Holz, Mario Fritz (from the arXiv API)", "dt": "2023-02-23 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2302.12173", "n": 344, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0343", "t": "LLMSecEval: A Dataset of Natural Language Prompts for Security Evaluations (arXiv 2303.09384)", "o": "Academic paper by Catherine Tony, Markus Mutas, Nicolás E. Díaz Ferreyra, Riccardo Scandariato (from the arXiv API)", "dt": "2023-03-16 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2303.09384", "n": 345, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0344", "t": "Those Aren't Your Memories, They're Somebody Else's: Seeding Misinformation in Chat Bot Memories (arXiv 2304.05371)", "o": "Academic paper by Conor Atkins, Benjamin Zi Hao Zhao, Hassan Jameel Asghar, Ian Wood, Mohamed Ali Kaafar (from the arXiv API)", "dt": "2023-04-06 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2304.05371", "n": 346, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0345", "t": "Poisoning Language Models During Instruction Tuning (arXiv 2305.00944)", "o": "Academic paper by Alexander Wan, Eric Wallace, Sheng Shen, Dan Klein (from the arXiv API)", "dt": "2023-05-01 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2305.00944", "n": 347, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0346", "t": "Sentence Embedding Leaks More Information than You Expect: Generative Embedding Inversion Attack to Recover the Whole Sentence (arXiv 2305.03010)", "o": "Academic paper by Haoran Li, Mingshi Xu, Yangqiu Song (from the arXiv API)", "dt": "2023-05-04 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2305.03010", "n": 348, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0347", "t": "Reconstructing Training Data from Multiclass Neural Networks (arXiv 2305.03350)", "o": "Academic paper by Gon Buzaglo, Niv Haim, Gilad Yehudai, Gal Vardi, Michal Irani (from the arXiv API)", "dt": "2023-05-05 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2305.03350", "n": 349, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0348", "t": "Privacy Auditing with One (1) Training Run (arXiv 2305.08846)", "o": "Academic paper by Thomas Steinke, Milad Nasr, Matthew Jagielski (from the arXiv API)", "dt": "2023-05-15 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2305.08846", "n": 350, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0349", "t": "Jailbreaking ChatGPT via Prompt Engineering: An Empirical Study (arXiv 2305.13860)", "o": "Academic paper by Yi Liu, Gelei Deng, Zhengzi Xu, Yuekang Li, Yaowen Zheng, Ying Zhang, and 4 others (from the arXiv API)", "dt": "2023-05-23 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2305.13860", "n": 351, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0350", "t": "Unleashing the Power of Randomization in Auditing Differentially Private ML (arXiv 2305.18447)", "o": "Academic paper by Krishna Pillutla, Galen Andrew, Peter Kairouz, H. Brendan McMahan, Alina Oprea, Sewoong Oh (from the arXiv API)", "dt": "2023-05-29 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2305.18447", "n": 352, "dec": "duplicate", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0351", "t": "Prompt Injection attack against LLM-integrated Applications (arXiv 2306.05499)", "o": "Academic paper by Yi Liu, Gelei Deng, Yuekang Li, Kailong Wang, Zihao Wang, Xiaofeng Wang, and 6 others (from the arXiv API)", "dt": "2023-06-08 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2306.05499", "n": 353, "dec": "duplicate", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0352", "t": "Explore, Establish, Exploit: Red Teaming Language Models from Scratch (arXiv 2306.09442)", "o": "Academic paper by Stephen Casper, Jason Lin, Joe Kwon, Gatlen Culp, Dylan Hadfield-Menell (from the arXiv API)", "dt": "2023-06-15 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2306.09442", "n": 356, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0353", "t": "Are aligned neural networks adversarially aligned? (arXiv 2306.15447)", "o": "Academic paper by Nicholas Carlini, Milad Nasr, Christopher A. Choquette-Choo, Matthew Jagielski, Irena Gao, Anas Awadalla, and 5 others (from the arXiv API)", "dt": "2023-06-26 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2306.15447", "n": 357, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0354", "t": "Jailbroken: How Does LLM Safety Training Fail? (arXiv 2307.02483)", "o": "Academic paper by Alexander Wei, Nika Haghtalab, Jacob Steinhardt (from the arXiv API)", "dt": "2023-07-05 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2307.02483", "n": 358, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0355", "t": "Effective Prompt Extraction from Language Models (arXiv 2307.06865)", "o": "Academic paper by Yiming Zhang, Nicholas Carlini, Daphne Ippolito (from the arXiv API)", "dt": "2023-07-13 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2307.06865", "n": 359, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0356", "t": "Rethinking Backdoor Attacks (arXiv 2307.10163)", "o": "Academic paper by Alaa Khaddaj, Guillaume Leclerc, Aleksandar Makelov, Kristian Georgiev, Hadi Salman, Andrew Ilyas, and 1 others (from the arXiv API)", "dt": "2023-07-19 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2307.10163", "n": 360, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0357", "t": "Abusing Images and Sounds for Indirect Instruction Injection in Multi-Modal LLMs (arXiv 2307.10490)", "o": "Academic paper by Eugene Bagdasaryan, Tsung-Yin Hsieh, Ben Nassi, Vitaly Shmatikov (from the arXiv API)", "dt": "2023-07-19 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2307.10490", "n": 361, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0358", "t": "LLM Censorship: A Machine Learning Challenge or a Computer Security Problem? (arXiv 2307.10719)", "o": "Academic paper by David Glukhov, Ilia Shumailov, Yarin Gal, Nicolas Papernot, Vardan Papyan (from the arXiv API)", "dt": "2023-07-20 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2307.10719", "n": 362, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0359", "t": "Universal and Transferable Adversarial Attacks on Aligned Language Models (arXiv 2307.15043)", "o": "Academic paper by Andy Zou, Zifan Wang, Nicholas Carlini, Milad Nasr, J. Zico Kolter, Matt Fredrikson (from the arXiv API)", "dt": "2023-07-27 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2307.15043", "n": 363, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0360", "t": "Backdooring Instruction-Tuned Large Language Models with Virtual Prompt Injection (arXiv 2307.16888)", "o": "Academic paper by Jun Yan, Vikas Yadav, Shiyang Li, Lichang Chen, Zheng Tang, Hai Wang, and 3 others (from the arXiv API)", "dt": "2023-07-31 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2307.16888", "n": 364, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0361", "t": "From Prompt Injections to SQL Injection Attacks: How Protected is Your LLM-Integrated Web Application? (arXiv 2308.01990)", "o": "Academic paper by Rodrigo Pedro, Daniel Castro, Paulo Carreira, Nuno Santos (from the arXiv API)", "dt": "2023-08-03 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2308.01990", "n": 365, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0362", "t": "\"Do Anything Now\": Characterizing and Evaluating In-The-Wild Jailbreak Prompts on Large Language Models (arXiv 2308.03825)", "o": "Academic paper by Xinyue Shen, Zeyuan Chen, Michael Backes, Yun Shen, Yang Zhang (from the arXiv API)", "dt": "2023-08-07 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2308.03825", "n": 366, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0363", "t": "Do-Not-Answer: A Dataset for Evaluating Safeguards in LLMs (arXiv 2308.13387)", "o": "Academic paper by Yuxia Wang, Haonan Li, Xudong Han, Preslav Nakov, Timothy Baldwin (from the arXiv API)", "dt": "2023-08-25 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2308.13387", "n": 367, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0364", "t": "Detecting Language Model Attacks with Perplexity (arXiv 2308.14132)", "o": "Academic paper by Gabriel Alon, Michael Kamfonas (from the arXiv API)", "dt": "2023-08-27 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2308.14132", "n": 368, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0365", "t": "Baseline Defenses for Adversarial Attacks Against Aligned Language Models (arXiv 2309.00614)", "o": "Academic paper by Neel Jain, Avi Schwarzschild, Yuxin Wen, Gowthami Somepalli, John Kirchenbauer, Ping-yeh Chiang, and 4 others (from the arXiv API)", "dt": "2023-09-01 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2309.00614", "n": 369, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0366", "t": "Dropout Attacks (arXiv 2309.01614)", "o": "Academic paper by Andrew Yuan, Alina Oprea, Cheng Tan (from the arXiv API)", "dt": "2023-09-04 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2309.01614", "n": 370, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0367", "t": "SmoothLLM: Defending Large Language Models Against Jailbreaking Attacks (arXiv 2310.03684)", "o": "Academic paper by Alexander Robey, Eric Wong, Hamed Hassani, George J. Pappas (from the arXiv API)", "dt": "2023-10-05 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2310.03684", "n": 371, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0368", "t": "Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To! (arXiv 2310.03693)", "o": "Academic paper by Xiangyu Qi, Yi Zeng, Tinghao Xie, Pin-Yu Chen, Ruoxi Jia, Prateek Mittal, and 1 others (from the arXiv API)", "dt": "2023-10-05 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2310.03693", "n": 372, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0369", "t": "Jailbreak and Guard Aligned Language Models with Only Few In-Context Demonstrations (arXiv 2310.06387)", "o": "Academic paper by Zeming Wei, Yifei Wang, Ang Li, Yichuan Mo, Yisen Wang (from the arXiv API)", "dt": "2023-10-10 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2310.06387", "n": 373, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0370", "t": "Text Embeddings Reveal (Almost) As Much As Text (arXiv 2310.06816)", "o": "Academic paper by John X. Morris, Volodymyr Kuleshov, Vitaly Shmatikov, Alexander M. Rush (from the arXiv API)", "dt": "2023-10-10 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2310.06816", "n": 374, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0371", "t": "LLMs Killed the Script Kiddie: How Agents Supported by Large Language Models Change the Landscape of Network Threat Testing (arXiv 2310.06936)", "o": "Academic paper by Stephen Moskal, Sam Laney, Erik Hemberg, Una-May O'Reilly (from the arXiv API)", "dt": "2023-10-10 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2310.06936", "n": 375, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0372", "t": "Beyond Memorization: Violating Privacy Via Inference with Large Language Models (arXiv 2310.07298)", "o": "Academic paper by Robin Staab, Mark Vero, Mislav Balunović, Martin Vechev (from the arXiv API)", "dt": "2023-10-11 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2310.07298", "n": 376, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0373", "t": "Deepfakes, Phrenology, Surveillance, and More! A Taxonomy of AI Privacy Risks (arXiv 2310.07879)", "o": "Academic paper by Hao-Ping Lee, Yu-Ju Yang, Thomas Serban von Davier, Jodi Forlizzi, Sauvik Das (from the arXiv API)", "dt": "2023-10-11 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2310.07879", "n": 377, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0374", "t": "Jailbreaking Black Box Large Language Models in Twenty Queries (arXiv 2310.08419)", "o": "Academic paper by Patrick Chao, Alexander Robey, Edgar Dobriban, Hamed Hassani, George J. Pappas, Eric Wong (from the arXiv API)", "dt": "2023-10-12 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2310.08419", "n": 378, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0375", "t": "NeMo Guardrails: A Toolkit for Controllable and Safe LLM Applications with Programmable Rails (arXiv 2310.10501)", "o": "Academic paper by Traian Rebedea, Razvan Dinu, Makesh Sreedhar, Christopher Parisien, Jonathan Cohen (from the arXiv API)", "dt": "2023-10-16 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2310.10501", "n": 379, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0376", "t": "AutoDAN: Interpretable Gradient-Based Adversarial Attacks on Large Language Models (arXiv 2310.15140)", "o": "Academic paper by Sicheng Zhu, Ruiyi Zhang, Bang An, Gang Wu, Joe Barrow, Zichao Wang, and 3 others (from the arXiv API)", "dt": "2023-10-23 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2310.15140", "n": 380, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0377", "t": "The Janus Interface: How Fine-Tuning in Large Language Models Amplifies the Privacy Risks (arXiv 2310.15469)", "o": "Academic paper by Xiaoyi Chen, Siyuan Tang, Rui Zhu, Shijun Yan, Lei Jin, Zihao Wang, and 4 others (from the arXiv API)", "dt": "2023-10-24 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2310.15469", "n": 381, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0378", "t": "Poisoning Retrieval Corpora by Injecting Adversarial Passages (arXiv 2310.19156)", "o": "Academic paper by Zexuan Zhong, Ziqing Huang, Alexander Wettig, Danqi Chen (from the arXiv API)", "dt": "2023-10-29 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2310.19156", "n": 382, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0379", "t": "BadLlama: cheaply removing safety fine-tuning from Llama 2-Chat 13B (arXiv 2311.00117)", "o": "Academic paper by Pranav Gade, Simon Lermen, Charlie Rogers-Smith, Jeffrey Ladish (from the arXiv API)", "dt": "2023-10-31 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2311.00117", "n": 383, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0380", "t": "Watermarks in the Sand: Impossibility of Strong Watermarking for Generative Models (arXiv 2311.04378)", "o": "Academic paper by Hanlin Zhang, Benjamin L. Edelman, Danilo Francati, Daniele Venturi, Giuseppe Ateniese, Boaz Barak (from the arXiv API)", "dt": "2023-11-07 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2311.04378", "n": 384, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0381", "t": "Removing RLHF Protections in GPT-4 via Fine-Tuning (arXiv 2311.05553)", "o": "Academic paper by Qiusi Zhan, Richard Fang, Rohan Bindu, Akul Gupta, Tatsunori Hashimoto, Daniel Kang (from the arXiv API)", "dt": "2023-11-09 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2311.05553", "n": 385, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0382", "t": "FigStep: Jailbreaking Large Vision-Language Models via Typographic Visual Prompts (arXiv 2311.05608)", "o": "Academic paper by Yichen Gong, Delong Ran, Jinyuan Liu, Conglei Wang, Tianshuo Cong, Anyu Wang, and 2 others (from the arXiv API)", "dt": "2023-11-09 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2311.05608", "n": 386, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0383", "t": "Token-Level Adversarial Prompt Detection Based on Perplexity Measures and Contextual Information (arXiv 2311.11509)", "o": "Academic paper by Zhengmian Hu, Gang Wu, Saayan Mitra, Ruiyi Zhang, Tong Sun, Heng Huang, and 1 others (from the arXiv API)", "dt": "2023-11-20 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2311.11509", "n": 387, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0384", "t": "Language Model Inversion (arXiv 2311.13647)", "o": "Academic paper by John X. Morris, Wenting Zhao, Justin T. Chiu, Vitaly Shmatikov, Alexander M. Rush (from the arXiv API)", "dt": "2023-11-22 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2311.13647", "n": 388, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0385", "t": "Universal Jailbreak Backdoors from Poisoned Human Feedback (arXiv 2311.14455)", "o": "Academic paper by Javier Rando, Florian Tramèr (from the arXiv API)", "dt": "2023-11-24 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2311.14455", "n": 389, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0386", "t": "Ignore This Title and HackAPrompt: Exposing Systemic Vulnerabilities of LLMs through a Global Scale Prompt Hacking Competition (arXiv 2311.16119)", "o": "Academic paper by Sander Schulhoff, Jeremy Pinto, Anaum Khan, Louis-François Bouchard, Chenglei Si, Svetlina Anati, and 4 others (from the arXiv API)", "dt": "2023-10-24 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2311.16119", "n": 390, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0387", "t": "Scalable Extraction of Training Data from (Production) Language Models (arXiv 2311.17035)", "o": "Academic paper by Milad Nasr, Nicholas Carlini, Jonathan Hayase, Matthew Jagielski, A. Feder Cooper, Daphne Ippolito, and 4 others (from the arXiv API)", "dt": "2023-11-28 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2311.17035", "n": 391, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0388", "t": "Tree of Attacks: Jailbreaking Black-Box LLMs Automatically (arXiv 2312.02119)", "o": "Academic paper by Anay Mehrotra, Manolis Zampetakis, Paul Kassianik, Blaine Nelson, Hyrum Anderson, Yaron Singer, and 1 others (from the arXiv API)", "dt": "2023-12-04 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2312.02119", "n": 392, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0389", "t": "Purple Llama CyberSecEval: A Secure Coding Benchmark for Language Models (arXiv 2312.04724)", "o": "Academic paper by Manish Bhatt, Sahana Chennabasappa, Cyrus Nikolaidis, Shengye Wan, Ivan Evtimov, Dominik Gabi, and 15 others (from the arXiv API)", "dt": "2023-12-07 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2312.04724", "n": 393, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0390", "t": "Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations (arXiv 2312.06674)", "o": "Academic paper by Hakan Inan, Kartikeya Upasani, Jianfeng Chi, Rashi Rungta, Krithika Iyer, Yuning Mao, and 5 others (from the arXiv API)", "dt": "2023-12-07 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2312.06674", "n": 394, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0391", "t": "The Adaptive Arms Race: Redefining Robustness in AI Security (arXiv 2312.13435)", "o": "Academic paper by Ilias Tsingenopoulos, Vera Rimmer, Davy Preuveneers, Fabio Pierazzi, Lorenzo Cavallaro, Wouter Joosen (from the arXiv API)", "dt": "2023-12-20 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2312.13435", "n": 395, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0392", "t": "Jatmo: Prompt Injection Defense by Task-Specific Finetuning (arXiv 2312.17673)", "o": "Academic paper by Julien Piet, Maha Alrashed, Chawin Sitawarin, Sizhe Chen, Zeming Wei, Elizabeth Sun, and 2 others (from the arXiv API)", "dt": "2023-12-29 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2312.17673", "n": 396, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0393", "t": "Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training (arXiv 2401.05566)", "o": "Academic paper by Evan Hubinger, Carson Denison, Jesse Mu, Mike Lambert, Meg Tong, Monte MacDiarmid, and 33 others (from the arXiv API)", "dt": "2024-01-10 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2401.05566", "n": 397, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0394", "t": "Red-Teaming for Generative AI: Silver Bullet or Security Theater? (arXiv 2401.15897)", "o": "Academic paper by Michael Feffer, Anusha Sinha, Wesley Hanwen Deng, Zachary C. Lipton, Hoda Heidari (from the arXiv API)", "dt": "2024-01-29 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2401.15897", "n": 398, "dec": "duplicate", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0395", "t": "An Early Categorization of Prompt Injection Attacks on Large Language Models (arXiv 2402.00898)", "o": "Academic paper by Sippo Rossi, Alisia Marianne Michel, Raghava Rao Mukkamala, Jason Bennett Thatcher (from the arXiv API)", "dt": "2024-01-31 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2402.00898", "n": 399, "dec": "duplicate", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0396", "t": "HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal (arXiv 2402.04249)", "o": "Academic paper by Mantas Mazeika, Long Phan, Xuwang Yin, Andy Zou, Zifan Wang, Norman Mu, and 6 others (from the arXiv API)", "dt": "2024-02-06 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2402.04249", "n": 402, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0397", "t": "LLM Agents can Autonomously Hack Websites (arXiv 2402.06664)", "o": "Academic paper by Richard Fang, Rohan Bindu, Akul Gupta, Qiusi Zhan, Daniel Kang (from the arXiv API)", "dt": "2024-02-06 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2402.06664", "n": 403, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0398", "t": "Architectural Neural Backdoors from First Principles (arXiv 2402.06957)", "o": "Academic paper by Harry Langford, Ilia Shumailov, Yiren Zhao, Robert Mullins, Nicolas Papernot (from the arXiv API)", "dt": "2024-02-10 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2402.06957", "n": 404, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0399", "t": "PoisonedRAG: Knowledge Corruption Attacks to Retrieval-Augmented Generation of Large Language Models (arXiv 2402.07867)", "o": "Academic paper by Wei Zou, Runpeng Geng, Binghui Wang, Jinyuan Jia (from the arXiv API)", "dt": "2024-02-12 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2402.07867", "n": 405, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0400", "t": "Soft Prompt Threats: Attacking Safety Alignment and Unlearning in Open-Source LLMs through the Embedding Space (arXiv 2402.09063)", "o": "Academic paper by Leo Schwinn, David Dobre, Sophie Xhonneux, Gauthier Gidel, Stephan Gunnemann (from the arXiv API)", "dt": "2024-02-14 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2402.09063", "n": 406, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0401", "t": "Instruction Tuning for Secure Code Generation (arXiv 2402.09497)", "o": "Academic paper by Jingxuan He, Mark Vero, Gabriela Krasnopolska, Martin Vechev (from the arXiv API)", "dt": "2024-02-14 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2402.09497", "n": 407, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0402", "t": "A StrongREJECT for Empty Jailbreaks (arXiv 2402.10260)", "o": "Academic paper by Alexandra Souly, Qingyuan Lu, Dillon Bowen, Tu Trinh, Elvis Hsieh, Sana Pandey, and 5 others (from the arXiv API)", "dt": "2024-02-15 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2402.10260", "n": 408, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0403", "t": "ArtPrompt: ASCII Art-based Jailbreak Attacks against Aligned LLMs (arXiv 2402.11753)", "o": "Academic paper by Fengqing Jiang, Zhangchen Xu, Luyao Niu, Zhen Xiang, Bhaskar Ramasubramanian, Bo Li, and 1 others (from the arXiv API)", "dt": "2024-02-19 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2402.11753", "n": 409, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0404", "t": "Query-Based Adversarial Prompt Generation (arXiv 2402.12329)", "o": "Academic paper by Jonathan Hayase, Ema Borevkovic, Nicholas Carlini, Florian Tramèr, Milad Nasr (from the arXiv API)", "dt": "2024-02-19 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2402.12329", "n": 410, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0405", "t": "Understanding and Mitigating the Threat of Vec2Text to Dense Retrieval Systems (arXiv 2402.12784)", "o": "Academic paper by Shengyao Zhuang, Bevan Koopman, Xiaoran Chu, Guido Zuccon (from the arXiv API)", "dt": "2024-02-20 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2402.12784", "n": 411, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0406", "t": "Fast Adversarial Attacks on Language Models In One GPU Minute (arXiv 2402.15570)", "o": "Academic paper by Vinu Sankar Sadasivan, Shoumik Saha, Gaurang Sriramanan, Priyatham Kattakinda, Atoosa Chegini, Soheil Feizi (from the arXiv API)", "dt": "2024-02-23 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2402.15570", "n": 412, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0407", "t": "PRP: Propagating Universal Perturbations to Attack Large Language Model Guard-Rails (arXiv 2402.15911)", "o": "Academic paper by Neal Mangaokar, Ashish Hooda, Jihye Choi, Shreyas Chandrashekaran, Kassem Fawaz, Somesh Jha, and 1 others (from the arXiv API)", "dt": "2024-02-24 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2402.15911", "n": 413, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0408", "t": "A New Era in LLM Security: Exploring Security Concerns in Real-World LLM-based Systems (arXiv 2402.18649)", "o": "Academic paper by Fangzhou Wu, Ning Zhang, Somesh Jha, Patrick McDaniel, Chaowei Xiao (from the arXiv API)", "dt": "2024-02-28 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2402.18649", "n": 414, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0409", "t": "Watermark Stealing in Large Language Models (arXiv 2402.19361)", "o": "Academic paper by Nikola Jovanović, Robin Staab, Martin Vechev (from the arXiv API)", "dt": "2024-02-29 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2402.19361", "n": 415, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0410", "t": "Teach LLMs to Phish: Stealing Private Information from Language Models (arXiv 2403.00871)", "o": "Academic paper by Ashwinee Panda, Christopher A. Choquette-Choo, Zhengming Zhang, Yaoqing Yang, Prateek Mittal (from the arXiv API)", "dt": "2024-03-01 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2403.00871", "n": 416, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0411", "t": "InjecAgent: Benchmarking Indirect Prompt Injections in Tool-Integrated Large Language Model Agents (arXiv 2403.02691)", "o": "Academic paper by Qiusi Zhan, Zhixiang Liang, Zifan Ying, Daniel Kang (from the arXiv API)", "dt": "2024-03-05 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2403.02691", "n": 417, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0412", "t": "Here Comes The AI Worm: Unleashing Zero-click Worms that Target GenAI-Powered Applications (arXiv 2403.02817)", "o": "Academic paper by Stav Cohen, Ron Bitton, Ben Nassi (from the arXiv API)", "dt": "2024-03-05 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2403.02817", "n": 418, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0413", "t": "The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearning (arXiv 2403.03218)", "o": "Academic paper by Nathaniel Li, Alexander Pan, Anjali Gopal, Summer Yue, Daniel Berrios, Alice Gatti, and 51 others (from the arXiv API)", "dt": "2024-03-05 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2403.03218", "n": 419, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0414", "t": "Neural Exec: Learning (and Learning from) Execution Triggers for Prompt Injection Attacks (arXiv 2403.03792)", "o": "Academic paper by Dario Pasquini, Martin Strohmeier, Carmela Troncoso (from the arXiv API)", "dt": "2024-03-06 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2403.03792", "n": 420, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0415", "t": "Automatic and Universal Prompt Injection Attacks against Large Language Models (arXiv 2403.04957)", "o": "Academic paper by Xiaogeng Liu, Zhiyuan Yu, Yizhe Zhang, Ning Zhang, Chaowei Xiao (from the arXiv API)", "dt": "2024-03-07 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2403.04957", "n": 421, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0416", "t": "IsolateGPT: An Execution Isolation Architecture for LLM-Based Agentic Systems (arXiv 2403.04960)", "o": "Academic paper by Yuhao Wu, Franziska Roesner, Tadayoshi Kohno, Ning Zhang, Umar Iqbal (from the arXiv API)", "dt": "2024-03-08 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2403.04960", "n": 422, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0417", "t": "Asset-centric Threat Modeling for AI-based Systems (arXiv 2403.06512)", "o": "Academic paper by Jan von der Assen, Jamo Sharif, Chao Feng, Christian Killer, Gérôme Bovet, Burkhard Stiller (from the arXiv API)", "dt": "2024-03-11 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2403.06512", "n": 423, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0418", "t": "Stealing Part of a Production Language Model (arXiv 2403.06634)", "o": "Academic paper by Nicholas Carlini, Daniel Paleka, Krishnamurthy Dj Dvijotham, Thomas Steinke, Jonathan Hayase, A. Feder Cooper, and 8 others (from the arXiv API)", "dt": "2024-03-11 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2403.06634", "n": 424, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0419", "t": "What Was Your Prompt? A Remote Keylogging Attack on AI Assistants (arXiv 2403.09751)", "o": "Academic paper by Roy Weiss, Daniel Ayzenshteyn, Guy Amit, Yisroel Mirsky (from the arXiv API)", "dt": "2024-03-14 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2403.09751", "n": 425, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0420", "t": "Defending Against Indirect Prompt Injection Attacks With Spotlighting (arXiv 2403.14720)", "o": "Academic paper by Keegan Hines, Gary Lopez, Matthew Hall, Federico Zarfati, Yonatan Zunger, Emre Kiciman (from the arXiv API)", "dt": "2024-03-20 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2403.14720", "n": 426, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0421", "t": "Don't Listen To Me: Understanding and Exploring Jailbreak Prompts of Large Language Models (arXiv 2403.17336)", "o": "Academic paper by Zhiyuan Yu, Xiaogeng Liu, Shunning Liang, Zach Cameron, Chaowei Xiao, Ning Zhang (from the arXiv API)", "dt": "2024-03-26 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2403.17336", "n": 427, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0422", "t": "JailbreakBench: An Open Robustness Benchmark for Jailbreaking Large Language Models (arXiv 2404.01318)", "o": "Academic paper by Patrick Chao, Edoardo Debenedetti, Alexander Robey, Maksym Andriushchenko, Francesco Croce, Vikash Sehwag, and 6 others (from the arXiv API)", "dt": "2024-03-28 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2404.01318", "n": 428, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0423", "t": "Great, Now Write an Article About That: The Crescendo Multi-Turn LLM Jailbreak Attack (arXiv 2404.01833)", "o": "Academic paper by Mark Russinovich, Ahmed Salem, Ronen Eldan (from the arXiv API)", "dt": "2024-04-02 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2404.01833", "n": 429, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0424", "t": "Jailbreaking Leading Safety-Aligned LLMs with Simple Adaptive Attacks (arXiv 2404.02151)", "o": "Academic paper by Maksym Andriushchenko, Francesco Croce, Nicolas Flammarion (from the arXiv API)", "dt": "2024-04-02 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2404.02151", "n": 430, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0425", "t": "LLM Agents can Autonomously Exploit One-day Vulnerabilities (arXiv 2404.08144)", "o": "Academic paper by Richard Fang, Rohan Bindu, Akul Gupta, Daniel Kang (from the arXiv API)", "dt": "2024-04-11 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2404.08144", "n": 431, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0426", "t": "FCert: Certifiably Robust Few-Shot Classification in the Era of Foundation Models (arXiv 2404.08631)", "o": "Academic paper by Yanting Wang, Wei Zou, Jinyuan Jia (from the arXiv API)", "dt": "2024-04-12 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2404.08631", "n": 432, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0427", "t": "CyberSecEval 2: A Wide-Ranging Cybersecurity Evaluation Suite for Large Language Models (arXiv 2404.13161)", "o": "Academic paper by Manish Bhatt, Sahana Chennabasappa, Yue Li, Cyrus Nikolaidis, Daniel Song, Shengye Wan, and 7 others (from the arXiv API)", "dt": "2024-04-19 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2404.13161", "n": 433, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0428", "t": "The Instruction Hierarchy: Training LLMs to Prioritize Privileged Instructions (arXiv 2404.13208)", "o": "Academic paper by Eric Wallace, Kai Xiao, Reimar Leike, Lilian Weng, Johannes Heidecke, Alex Beutel (from the arXiv API)", "dt": "2024-04-19 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2404.13208", "n": 434, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0429", "t": "AirGapAgent: Protecting Privacy-Conscious Conversational Agents (arXiv 2405.05175)", "o": "Academic paper by Eugene Bagdasarian, Ren Yi, Sahra Ghalebikesabi, Peter Kairouz, Marco Gruteser, Sewoong Oh, and 2 others (from the arXiv API)", "dt": "2024-05-08 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2405.05175", "n": 435, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0430", "t": "PLeak: Prompt Leaking Attacks against Large Language Model Applications (arXiv 2405.06823)", "o": "Academic paper by Bo Hui, Haolin Yuan, Neil Gong, Philippe Burlina, Yinzhi Cao (from the arXiv API)", "dt": "2024-05-10 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2405.06823", "n": 436, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0431", "t": "Exploiting LLM Quantization (arXiv 2405.18137)", "o": "Academic paper by Kazuki Egashira, Mark Vero, Robin Staab, Jingxuan He, Martin Vechev (from the arXiv API)", "dt": "2024-05-28 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2405.18137", "n": 437, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0432", "t": "Phantom: General Backdoor Attacks on Retrieval Augmented Language Generation (arXiv 2405.20485)", "o": "Academic paper by Harsh Chaudhari, Giorgio Severi, John Abascal, Anshuman Suri, Matthew Jagielski, Christopher A. Choquette-Choo, and 3 others (from the arXiv API)", "dt": "2024-05-30 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2405.20485", "n": 438, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0433", "t": "BadRAG: Identifying Vulnerabilities in Retrieval Augmented Generation of Large Language Models (arXiv 2406.00083)", "o": "Academic paper by Jiaqi Xue, Mengxin Zheng, Yebowen Hu, Fei Liu, Xun Chen, Qian Lou (from the arXiv API)", "dt": "2024-06-03 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2406.00083", "n": 439, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0434", "t": "Improving Alignment and Robustness with Circuit Breakers (arXiv 2406.04313)", "o": "Academic paper by Andy Zou, Long Phan, Justin Wang, Derek Duenas, Maxwell Lin, Maksym Andriushchenko, and 4 others (from the arXiv API)", "dt": "2024-06-06 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2406.04313", "n": 440, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0435", "t": "Security of AI Agents (arXiv 2406.08689)", "o": "Academic paper by Yifeng He, Ethan Wang, Yuyang Rong, Zifei Cheng, Hao Chen (from the arXiv API)", "dt": "2024-06-12 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2406.08689", "n": 441, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0436", "t": "GuardAgent: Safeguard LLM Agents by a Guard Agent via Knowledge-Enabled Reasoning (arXiv 2406.09187)", "o": "Academic paper by Zhen Xiang, Linzhi Zheng, Yanjie Li, Junyuan Hong, Qinbin Li, Han Xie, and 6 others (from the arXiv API)", "dt": "2024-06-13 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2406.09187", "n": 442, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0437", "t": "We Have a Package for You! A Comprehensive Analysis of Package Hallucinations by Code Generating LLMs (arXiv 2406.10279)", "o": "Academic paper by Joseph Spracklen, Raveen Wijewickrama, A H M Nazmus Sakib, Anindya Maiti, Bimal Viswanath, Murtuza Jadliwala (from the arXiv API)", "dt": "2024-06-12 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2406.10279", "n": 443, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0438", "t": "garak: A Framework for Security Probing Large Language Models (arXiv 2406.11036)", "o": "Academic paper by Leon Derczynski, Erick Galinkin, Jeffrey Martin, Subho Majumdar, Nanna Inie (from the arXiv API)", "dt": "2024-06-16 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2406.11036", "n": 444, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0439", "t": "Refusal in Language Models Is Mediated by a Single Direction (arXiv 2406.11717)", "o": "Academic paper by Andy Arditi, Oscar Obeso, Aaquib Syed, Daniel Paleka, Nina Panickssery, Wes Gurnee, and 1 others (from the arXiv API)", "dt": "2024-06-17 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2406.11717", "n": 445, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0440", "t": "IDs for AI Systems (arXiv 2406.12137)", "o": "Academic paper by Alan Chan, Noam Kolt, Peter Wills, Usman Anwar, Christian Schroeder de Witt, Nitarshan Rajkumar, and 4 others (from the arXiv API)", "dt": "2024-06-17 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2406.12137", "n": 446, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0441", "t": "Dissecting Adversarial Robustness of Multimodal LM Agents (arXiv 2406.12814)", "o": "Academic paper by Chen Henry Wu, Rishi Shah, Jing Yu Koh, Ruslan Salakhutdinov, Daniel Fried, Aditi Raghunathan (from the arXiv API)", "dt": "2024-06-18 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2406.12814", "n": 447, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0442", "t": "Current state of LLM Risks and AI Guardrails (arXiv 2406.12934)", "o": "Academic paper by Suriya Ganesh Ayyamperumal, Limin Ge (from the arXiv API)", "dt": "2024-06-16 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2406.12934", "n": 448, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0443", "t": "AgentDojo: A Dynamic Environment to Evaluate Prompt Injection Attacks and Defenses for LLM Agents (arXiv 2406.13352)", "o": "Academic paper by Edoardo Debenedetti, Jie Zhang, Mislav Balunović, Luca Beurer-Kellner, Marc Fischer, Florian Tramèr (from the arXiv API)", "dt": "2024-06-19 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2406.13352", "n": 449, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0444", "t": "WildGuard: Open One-Stop Moderation Tools for Safety Risks, Jailbreaks, and Refusals of LLMs (arXiv 2406.18495)", "o": "Academic paper by Seungju Han, Kavel Rao, Allyson Ettinger, Liwei Jiang, Bill Yuchen Lin, Nathan Lambert, and 2 others (from the arXiv API)", "dt": "2024-06-26 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2406.18495", "n": 450, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0445", "t": "Covert Malicious Finetuning: Challenges in Safeguarding LLM Adaptation (arXiv 2406.20053)", "o": "Academic paper by Danny Halawi, Alexander Wei, Eric Wallace, Tony T. Wang, Nika Haghtalab, Jacob Steinhardt (from the arXiv API)", "dt": "2024-06-28 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2406.20053", "n": 451, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0446", "t": "$R^2$-Guard: Robust Reasoning Enabled LLM Guardrail via Knowledge-Enhanced Logical Reasoning (arXiv 2407.05557)", "o": "Academic paper by Mintong Kang, Bo Li (from the arXiv API)", "dt": "2024-07-08 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2407.05557", "n": 452, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0447", "t": "Prompting Techniques for Secure Code Generation: A Systematic Investigation (arXiv 2407.07064)", "o": "Academic paper by Catherine Tony, Nicolás E. Díaz Ferreyra, Markus Mutas, Salem Dhiff, Riccardo Scandariato (from the arXiv API)", "dt": "2024-07-09 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2407.07064", "n": 453, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0448", "t": "AgentPoison: Red-teaming LLM Agents via Poisoning Memory or Knowledge Bases (arXiv 2407.12784)", "o": "Academic paper by Zhaorun Chen, Zhen Xiang, Chaowei Xiao, Dawn Song, Bo Li (from the arXiv API)", "dt": "2024-07-17 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2407.12784", "n": 454, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0449", "t": "Latent Adversarial Training Improves Robustness to Persistent Harmful Behaviors in LLMs (arXiv 2407.15549)", "o": "Academic paper by Abhay Sheshadri, Aidan Ewart, Phillip Guo, Aengus Lynch, Cindy Wu, Vivek Hebbar, and 5 others (from the arXiv API)", "dt": "2024-07-22 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2407.15549", "n": 455, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0450", "t": "FLRT: Fluent Student-Teacher Redteaming (arXiv 2407.17447)", "o": "Academic paper by T. Ben Thompson, Michael Sklar (from the arXiv API)", "dt": "2024-07-24 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2407.17447", "n": 456, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0451", "t": "The Emerged Security and Privacy of LLM Agent: A Survey with Case Studies (arXiv 2407.19354)", "o": "Academic paper by Feng He, Tianqing Zhu, Dayong Ye, Bo Liu, Wanlei Zhou, Philip S. Yu (from the arXiv API)", "dt": "2024-07-28 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2407.19354", "n": 457, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0452", "t": "Breaking Agents: Compromising Autonomous LLM Agents Through Malfunction Amplification (arXiv 2407.20859)", "o": "Academic paper by Boyang Zhang, Yicong Tan, Yun Shen, Ahmed Salem, Michael Backes, Savvas Zannettou, and 1 others (from the arXiv API)", "dt": "2024-07-30 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2407.20859", "n": 458, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0453", "t": "ShieldGemma: Generative AI Content Moderation Based on Gemma (arXiv 2407.21772)", "o": "Academic paper by Wenjun Zeng, Yuchi Liu, Ryan Mullins, Ludovic Peran, Joe Fernandez, Hamza Harkous, and 6 others (from the arXiv API)", "dt": "2024-07-31 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2407.21772", "n": 459, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0454", "t": "On the Resilience of LLM-Based Multi-Agent Collaboration with Faulty Agents (arXiv 2408.00989)", "o": "Academic paper by Jen-tse Huang, Jiaxu Zhou, Tailin Jin, Xuhui Zhou, Zixi Chen, Wenxuan Wang, and 3 others (from the arXiv API)", "dt": "2024-08-02 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2408.00989", "n": 460, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0455", "t": "CYBERSECEVAL 3: Advancing the Evaluation of Cybersecurity Risks and Capabilities in Large Language Models (arXiv 2408.01605)", "o": "Academic paper by Shengye Wan, Cyrus Nikolaidis, Daniel Song, David Molnar, James Crnkovich, Jayson Grace, and 7 others (from the arXiv API)", "dt": "2024-08-02 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2408.01605", "n": 461, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0456", "t": "Scaling Trends for Data Poisoning in LLMs (arXiv 2408.02946)", "o": "Academic paper by Dillon Bowen, Brendan Murphy, Will Cai, David Khachaturov, Adam Gleave, Kellin Pelrine (from the arXiv API)", "dt": "2024-08-06 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2408.02946", "n": 462, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0457", "t": "MPC-Minimized Secure LLM Inference (arXiv 2408.03561)", "o": "Academic paper by Deevashwer Rathee, Dacheng Li, Ion Stoica, Hao Zhang, Raluca Popa (from the arXiv API)", "dt": "2024-08-07 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2408.03561", "n": 463, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0458", "t": "Cybench: A Framework for Evaluating Cybersecurity Capabilities and Risks of Language Models (arXiv 2408.08926)", "o": "Academic paper by Andy K. Zhang, Neil Perry, Riya Dulepet, Joey Ji, Celeste Menders, Justin W. Lin, and 21 others (from the arXiv API)", "dt": "2024-08-15 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2408.08926", "n": 464, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0459", "t": "CLIBE: Detecting Dynamic Backdoors in Transformer-based NLP Models (arXiv 2409.01193)", "o": "Academic paper by Rui Zeng, Xi Chen, Yuwen Pu, Xuhong Zhang, Tianyu Du, Shouling Ji (from the arXiv API)", "dt": "2024-09-02 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2409.01193", "n": 465, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0460", "t": "EIA: Environmental Injection Attack on Generalist Web Agents for Privacy Leakage (arXiv 2409.11295)", "o": "Academic paper by Zeyi Liao, Lingbo Mo, Chejian Xu, Mintong Kang, Jiawei Zhang, Chaowei Xiao, and 3 others (from the arXiv API)", "dt": "2024-09-17 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2409.11295", "n": 466, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0461", "t": "Extracting Memorized Training Data via Decomposition (arXiv 2409.12367)", "o": "Academic paper by Ellen Su, Anu Vellore, Amy Chang, Raffaele Mura, Blaine Nelson, Paul Kassianik, and 1 others (from the arXiv API)", "dt": "2024-09-18 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2409.12367", "n": 467, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0462", "t": "On the Feasibility of Fully AI-automated Vishing Attacks (arXiv 2409.13793)", "o": "Academic paper by João Figueiredo, Afonso Carvalho, Daniel Castro, Daniel Gonçalves, Nuno Santos (from the arXiv API)", "dt": "2024-09-20 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2409.13793", "n": 468, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0463", "t": "Artificial-Intelligence Generated Code Considered Harmful: A Road Map for Secure and High-Quality Code Generation (arXiv 2409.19182)", "o": "Academic paper by Chun Jie Chong, Zhihao Yao, Iulian Neamtiu (from the arXiv API)", "dt": "2024-09-27 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2409.19182", "n": 469, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0464", "t": "Adversarial Decoding: Generating Readable Documents for Adversarial Objectives (arXiv 2410.02163)", "o": "Academic paper by Collin Zhang, Tingwei Zhang, Vitaly Shmatikov (from the arXiv API)", "dt": "2024-10-03 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2410.02163", "n": 470, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0465", "t": "Permissive Information-Flow Analysis for Large Language Models (arXiv 2410.03055)", "o": "Academic paper by Shoaib Ahmed Siddiqui, Radhika Gaonkar, Boris Köpf, David Krueger, Andrew Paverd, Ahmed Salem, and 4 others (from the arXiv API)", "dt": "2024-10-04 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2410.03055", "n": 471, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0466", "t": "SecAlign: Defending Against Prompt Injection with Preference Optimization (arXiv 2410.05451)", "o": "Academic paper by Sizhe Chen, Arman Zharmagambetov, Saeed Mahloujifar, Kamalika Chaudhuri, David Wagner, Chuan Guo (from the arXiv API)", "dt": "2024-10-07 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2410.05451", "n": 472, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0467", "t": "Prompt Infection: LLM-to-LLM Prompt Injection within Multi-Agent Systems (arXiv 2410.07283)", "o": "Academic paper by Donghyun Lee, Mo Tiwari (from the arXiv API)", "dt": "2024-10-09 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2410.07283", "n": 473, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0468", "t": "AgentHarm: A Benchmark for Measuring Harmfulness of LLM Agents (arXiv 2410.09024)", "o": "Academic paper by Maksym Andriushchenko, Alexandra Souly, Mateusz Dziemian, Derek Duenas, Maxwell Lin, Justin Wang, and 8 others (from the arXiv API)", "dt": "2024-10-11 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2410.09024", "n": 474, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0469", "t": "Catastrophic Cyber Capabilities Benchmark (3CB): Robustly Evaluating LLM Agent Cyber Offense Capabilities (arXiv 2410.09114)", "o": "Academic paper by Andrey Anurin, Jonathan Ng, Kibo Schaffer, Jason Schreiber, Esben Kran (from the arXiv API)", "dt": "2024-10-10 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2410.09114", "n": 475, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0470", "t": "Denial-of-Service Poisoning Attacks against Large Language Models (arXiv 2410.10760)", "o": "Academic paper by Kuofeng Gao, Tianyu Pang, Chao Du, Yong Yang, Shu-Tao Xia, Min Lin (from the arXiv API)", "dt": "2024-10-14 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2410.10760", "n": 476, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0471", "t": "SeCodePLT: A Unified Platform for Evaluating the Security of Code GenAI (arXiv 2410.11096)", "o": "Academic paper by Yuzhou Nie, Zhun Wang, Yu Yang, Ruizhe Jiang, Yuheng Tang, Xander Davies, and 4 others (from the arXiv API)", "dt": "2024-10-14 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2410.11096", "n": 477, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0472", "t": "Backdoored Retrievers for Prompt Injection Attacks on Retrieval Augmented Generation of Large Language Models (arXiv 2410.14479)", "o": "Academic paper by Cody Clop, Yannick Teglia (from the arXiv API)", "dt": "2024-10-18 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2410.14479", "n": 478, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0473", "t": "Imprompter: Tricking LLM Agents into Improper Tool Use (arXiv 2410.14923)", "o": "Academic paper by Xiaohan Fu, Shuheng Li, Zihan Wang, Yihao Liu, Rajesh K. Gupta, Taylor Berg-Kirkpatrick, and 1 others (from the arXiv API)", "dt": "2024-10-19 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2410.14923", "n": 479, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0474", "t": "AdvAgent: Controllable Blackbox Red-teaming on Web Agents (arXiv 2410.17401)", "o": "Academic paper by Chejian Xu, Mintong Kang, Jiawei Zhang, Zeyi Liao, Lingbo Mo, Mengqi Yuan, and 2 others (from the arXiv API)", "dt": "2024-10-22 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2410.17401", "n": 480, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0475", "t": "HonestCyberEval: An AI Cyber Risk Benchmark for Automated Software Exploitation (arXiv 2410.21939)", "o": "Academic paper by Dan Ristea, Vasilios Mavroudis (from the arXiv API)", "dt": "2024-10-29 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2410.21939", "n": 481, "dec": "duplicate", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0476", "t": "Attention Tracker: Detecting Prompt Injection Attacks in LLMs (arXiv 2411.00348)", "o": "Academic paper by Kuo-Han Hung, Ching-Yun Ko, Ambrish Rawat, I-Hsin Chung, Winston H. Hsu, Pin-Yu Chen (from the arXiv API)", "dt": "2024-11-01 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2411.00348", "n": 482, "dec": "duplicate", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0477", "t": "Attacking Vision-Language Computer Agents via Pop-ups (arXiv 2411.02391)", "o": "Academic paper by Yanzhe Zhang, Tao Yu, Diyi Yang (from the arXiv API)", "dt": "2024-11-04 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2411.02391", "n": 485, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0478", "t": "RedCode: Risky Code Execution and Generation Benchmark for Code Agents (arXiv 2411.07781)", "o": "Academic paper by Chengquan Guo, Xun Liu, Chulin Xie, Andy Zhou, Yi Zeng, Zinan Lin, and 2 others (from the arXiv API)", "dt": "2024-11-12 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2411.07781", "n": 486, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0479", "t": "Feedback-Guided Extraction of Knowledge Base from Retrieval-Augmented LLM Applications (arXiv 2411.14110)", "o": "Academic paper by Changyue Jiang, Xudong Pan, Geng Hong, Chenfu Bao, Yang Chen, Min Yang (from the arXiv API)", "dt": "2024-11-21 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2411.14110", "n": 487, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0480", "t": "Best-of-N Jailbreaking (arXiv 2412.03556)", "o": "Academic paper by John Hughes, Sara Price, Aengus Lynch, Rylan Schaeffer, Fazl Barez, Sanmi Koyejo, and 4 others (from the arXiv API)", "dt": "2024-12-04 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2412.03556", "n": 488, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0481", "t": "LeakAgent: RL-based Red-teaming Agent for LLM Privacy Leakage (arXiv 2412.05734)", "o": "Academic paper by Yuzhou Nie, Zhun Wang, Ye Yu, Xian Wu, Xuandong Zhao, Wenbo Guo, and 1 others (from the arXiv API)", "dt": "2024-12-07 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2412.05734", "n": 489, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0482", "t": "SafeWatch: An Efficient Safety-Policy Following Video Guardrail Model with Transparent Explanations (arXiv 2412.06878)", "o": "Academic paper by Zhaorun Chen, Francesco Pinto, Minzhou Pan, Bo Li (from the arXiv API)", "dt": "2024-12-09 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2412.06878", "n": 490, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0483", "t": "Agent-SafetyBench: Evaluating the Safety of LLM Agents (arXiv 2412.14470)", "o": "Academic paper by Zhexin Zhang, Shiyao Cui, Yida Lu, Jingzhuo Zhou, Junxiao Yang, Hongning Wang, and 1 others (from the arXiv API)", "dt": "2024-12-19 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2412.14470", "n": 491, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0484", "t": "The Task Shield: Enforcing Task Alignment to Defend Against Indirect Prompt Injection in LLM Agents (arXiv 2412.16682)", "o": "Academic paper by Feiran Jia, Tong Wu, Xin Qin, Anna Squicciarini (from the arXiv API)", "dt": "2024-12-21 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2412.16682", "n": 492, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0485", "t": "Lessons From Red Teaming 100 Generative AI Products (arXiv 2501.07238)", "o": "Academic paper by Blake Bullwinkel, Amanda Minnich, Shiven Chawla, Gary Lopez, Martin Pouliot, Whitney Maxwell, and 20 others (from the arXiv API)", "dt": "2025-01-13 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2501.07238", "n": 493, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0486", "t": "Fun-tuning: Characterizing the Vulnerability of Proprietary LLMs to Optimization-based Prompt Injection Attacks via the Fine-Tuning Interface (arXiv 2501.09798)", "o": "Academic paper by Andrey Labunets, Nishit V. Pandya, Ashish Hooda, Xiaohan Fu, Earlence Fernandes (from the arXiv API)", "dt": "2025-01-16 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2501.09798", "n": 494, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0487", "t": "Poison-RAG: Adversarial Data Poisoning Attacks on Retrieval-Augmented Generation in Recommender Systems (arXiv 2501.11759)", "o": "Academic paper by Fatemeh Nazary, Yashar Deldjoo, Tommaso di Noia (from the arXiv API)", "dt": "2025-01-20 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2501.11759", "n": 495, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0488", "t": "The Relationship Between Network Similarity and Transferability of Adversarial Attacks (arXiv 2501.18629)", "o": "Academic paper by Gerrit Klause, Niklas Bunzel (from the arXiv API)", "dt": "2025-01-27 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2501.18629", "n": 496, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0489", "t": "Constitutional Classifiers: Defending against Universal Jailbreaks across Thousands of Hours of Red Teaming (arXiv 2501.18837)", "o": "Academic paper by Mrinank Sharma, Meg Tong, Jesse Mu, Jerry Wei, Jorrit Kruthoff, Scott Goodfriend, and 37 others (from the arXiv API)", "dt": "2025-01-31 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2501.18837", "n": 497, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0490", "t": "LLM Cyber Evaluations Don't Capture Real-World Risk (arXiv 2502.00072)", "o": "Academic paper by Kamilė Lukošiūtė, Adam Swanda (from the arXiv API)", "dt": "2025-01-31 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2502.00072", "n": 498, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0491", "t": "MELON: Provable Defense Against Indirect Prompt Injection Attacks in AI Agents (arXiv 2502.05174)", "o": "Academic paper by Kaijie Zhu, Xianjun Yang, Jindong Wang, Wenbo Guo, William Yang Wang (from the arXiv API)", "dt": "2025-02-07 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2502.05174", "n": 499, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0492", "t": "Commercial LLM Agents Are Already Vulnerable to Simple Yet Dangerous Attacks (arXiv 2502.08586)", "o": "Academic paper by Ang Li, Yin Zhou, Vethavikashini Chithrra Raghuram, Tom Goldstein, Micah Goldblum (from the arXiv API)", "dt": "2025-02-12 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2502.08586", "n": 500, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0493", "t": "BaxBench: Can LLMs Generate Correct and Secure Backends? (arXiv 2502.11844)", "o": "Academic paper by Mark Vero, Niels Mündler, Victor Chibotaru, Veselin Raychev, Maximilian Baader, Nikola Jovanović, and 2 others (from the arXiv API)", "dt": "2025-02-17 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2502.11844", "n": 501, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0494", "t": "Evaluating the Robustness of Multimodal Agents Against Active Environmental Injection Attacks (arXiv 2502.13053)", "o": "Academic paper by Yurun Chen, Xavier Hu, Keting Yin, Juncheng Li, Shengyu Zhang (from the arXiv API)", "dt": "2025-02-18 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2502.13053", "n": 502, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0495", "t": "Red-Teaming LLM Multi-Agent Systems via Communication Attacks (arXiv 2502.14847)", "o": "Academic paper by Pengfei He, Yupin Lin, Shen Dong, Han Xu, Yue Xing, Hui Liu (from the arXiv API)", "dt": "2025-02-20 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2502.14847", "n": 503, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0496", "t": "OCCULT: Evaluating Large Language Models for Offensive Cyber Operation Capabilities (arXiv 2502.15797)", "o": "Academic paper by Michael Kouremetis, Marissa Dotter, Alex Byrne, Dan Martin, Ethan Michalak, Gianpaolo Russo, and 2 others (from the arXiv API)", "dt": "2025-02-18 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2502.15797", "n": 504, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0497", "t": "Can Indirect Prompt Injection Attacks Be Detected and Removed? (arXiv 2502.16580)", "o": "Academic paper by Yulin Chen, Haoran Li, Yuan Sui, Yufei He, Yue Liu, Yangqiu Song, and 1 others (from the arXiv API)", "dt": "2025-02-23 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2502.16580", "n": 505, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0498", "t": "PrivaCI-Bench: Evaluating Privacy with Contextual Integrity and Legal Compliance (arXiv 2502.17041)", "o": "Academic paper by Haoran Li, Wenbin Hu, Huihao Jing, Yulin Chen, Qi Hu, Sirui Han, and 3 others (from the arXiv API)", "dt": "2025-02-24 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2502.17041", "n": 506, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0499", "t": "MM-PoisonRAG: Disrupting Multimodal RAG with Local and Global Poisoning Attacks (arXiv 2502.17832)", "o": "Academic paper by Hyeonjeong Ha, Qiusi Zhan, Jeonghwan Kim, Dimitrios Bralios, Saikrishna Sanniboina, Nanyun Peng, and 3 others (from the arXiv API)", "dt": "2025-02-25 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2502.17832", "n": 507, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0500", "t": "SOK: Exploring Hallucinations and Security Risks in AI-Assisted Software Development with Insights for LLM Deployment (arXiv 2502.18468)", "o": "Academic paper by Ariful Haque, Sunzida Siddique, Md. Mahfuzur Rahman, Ahmed Rafi Hasan, Laxmi Rani Das, Marufa Kamal, and 2 others (from the arXiv API)", "dt": "2025-01-31 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2502.18468", "n": 508, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0501", "t": "Memory Injection Attacks on LLM Agents via Query-Only Interaction (arXiv 2503.03704)", "o": "Academic paper by Shen Dong, Shaochen Xu, Pengfei He, Yige Li, Jiliang Tang, Tianming Liu, and 2 others (from the arXiv API)", "dt": "2025-03-05 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2503.03704", "n": 509, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0502", "t": "Poisoned-MRAG: Knowledge Poisoning Attacks to Multimodal Retrieval Augmented Generation (arXiv 2503.06254)", "o": "Academic paper by Yinuo Liu, Zenghui Yuan, Guiyao Tie, Jiawen Shi, Pan Zhou, Lichao Sun, and 1 others (from the arXiv API)", "dt": "2025-03-08 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2503.06254", "n": 510, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0503", "t": "MIP against Agent: Malicious Image Patches Hijacking Multimodal OS Agents (arXiv 2503.10809)", "o": "Academic paper by Lukas Aichberger, Alasdair Paren, Guohao Li, Philip Torr, Yarin Gal, Adel Bibi (from the arXiv API)", "dt": "2025-03-13 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2503.10809", "n": 511, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0504", "t": "Prompt Injection Detection and Mitigation via AI Multi-Agent NLP Frameworks (arXiv 2503.11517)", "o": "Academic paper by Diego Gosmar, Deborah A. Dahl, Dario Gosmar (from the arXiv API)", "dt": "2025-03-14 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2503.11517", "n": 512, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0505", "t": "A Framework for Evaluating Emerging Cyberattack Capabilities of AI (arXiv 2503.11917)", "o": "Academic paper by Mikel Rodriguez, Raluca Ada Popa, Four Flynn, Lihao Liang, Allan Dafoe, Anna Wang (from the arXiv API)", "dt": "2025-03-14 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2503.11917", "n": 513, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0506", "t": "Multi-Agent Systems Execute Arbitrary Malicious Code (arXiv 2503.12188)", "o": "Academic paper by Harold Triedman, Rishi Jha, Vitaly Shmatikov (from the arXiv API)", "dt": "2025-03-15 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2503.12188", "n": 514, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0507", "t": "Rethinking the Evaluation of Secure Code Generation (arXiv 2503.15554)", "o": "Academic paper by Shih-Chieh Dai, Jun Xu, Guanhong Tao (from the arXiv API)", "dt": "2025-03-18 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2503.15554", "n": 515, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0508", "t": "CVE-Bench: A Benchmark for AI Agents' Ability to Exploit Real-World Web Application Vulnerabilities (arXiv 2503.17332)", "o": "Academic paper by Yuxuan Zhu, Antony Kellermann, Dylan Bowman, Philip Li, Akul Gupta, Adarsh Danda, and 10 others (from the arXiv API)", "dt": "2025-03-21 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2503.17332", "n": 516, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0509", "t": "AgentSpec: Customizable Runtime Enforcement for Safe and Reliable LLM Agents (arXiv 2503.18666)", "o": "Academic paper by Haoyu Wang, Christopher M. Poskitt, Jun Sun (from the arXiv API)", "dt": "2025-03-24 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2503.18666", "n": 517, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0510", "t": "Defeating Prompt Injections by Design (arXiv 2503.18813)", "o": "Academic paper by Edoardo Debenedetti, Ilia Shumailov, Tianqi Fan, Jamie Hayes, Nicholas Carlini, Daniel Fabian, and 4 others (from the arXiv API)", "dt": "2025-03-24 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2503.18813", "n": 518, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0511", "t": "sudo rm -rf agentic_security (arXiv 2503.20279)", "o": "Academic paper by Sejin Lee, Jian Kim, Haon Park, Ashkan Yousefpour, Sangyoon Yu, Min Song (from the arXiv API)", "dt": "2025-03-26 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2503.20279", "n": 519, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0512", "t": "ShieldAgent: Shielding Agents via Verifiable Safety Policy Reasoning (arXiv 2503.22738)", "o": "Academic paper by Zhaorun Chen, Mintong Kang, Bo Li (from the arXiv API)", "dt": "2025-03-26 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2503.22738", "n": 520, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0513", "t": "DrunkAgent: Stealthy Memory Corruption in LLM-Powered Recommender Agents (arXiv 2503.23804)", "o": "Academic paper by Shiyi Yang, Zhibo Hu, Xinshu Li, Chen Wang, Tong Yu, Xiwei Xu, and 2 others (from the arXiv API)", "dt": "2025-03-31 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2503.23804", "n": 521, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0514", "t": "SandboxEval: Towards Securing Test Environment for Untrusted Code (arXiv 2504.00018)", "o": "Academic paper by Rafiqul Rabin, Jesse Hostetler, Sean McGregor, Brett Weir, Nick Judd (from the arXiv API)", "dt": "2025-03-27 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2504.00018", "n": 522, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0515", "t": "Universal Zero-shot Embedding Inversion (arXiv 2504.00147)", "o": "Academic paper by Collin Zhang, John X. Morris, Vitaly Shmatikov (from the arXiv API)", "dt": "2025-03-31 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2504.00147", "n": 523, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0516", "t": "One Pic is All it Takes: Poisoning Visual Document Retrieval Augmented Generation with a Single Image (arXiv 2504.02132)", "o": "Academic paper by Ezzeldin Shereen, Dan Ristea, Shae McFadden, Burak Hasircioglu, Vasilios Mavroudis, Chris Hicks (from the arXiv API)", "dt": "2025-04-02 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2504.02132", "n": 524, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0517", "t": "MCP Safety Audit: LLMs with the Model Context Protocol Allow Major Security Exploits (arXiv 2504.03767)", "o": "Academic paper by Brandon Radosevich, John Halloran (from the arXiv API)", "dt": "2025-04-02 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2504.03767", "n": 525, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0518", "t": "Practical Poisoning Attacks against Retrieval-Augmented Generation (arXiv 2504.03957)", "o": "Academic paper by Baolei Zhang, Yuxi Chen, Zhuqing Liu, Lihai Nie, Tong Li, Zheli Liu, and 1 others (from the arXiv API)", "dt": "2025-04-04 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2504.03957", "n": 526, "dec": "duplicate", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0519", "t": "Bypassing LLM Guardrails: An Empirical Analysis of Evasion Attacks against Prompt Injection and Jailbreak Detection Systems (arXiv 2504.11168)", "o": "Academic paper by William Hackett, Lewis Birch, Stefan Trawicki, Neeraj Suri, Peter Garraghan (from the arXiv API)", "dt": "2025-04-15 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2504.11168", "n": 527, "dec": "duplicate", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0520", "t": "Manipulating Multimodal Agents via Cross-Modal Prompt Injection (arXiv 2504.14348)", "o": "Academic paper by Le Wang, Zonghao Ying, Tianyuan Zhang, Siyuan Liang, Shengshan Hu, Mingchuan Zhang, and 2 others (from the arXiv API)", "dt": "2025-04-19 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2504.14348", "n": 530, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0521", "t": "Breaking the Prompt Wall (I): A Real-World Case Study of Attacking ChatGPT via Lightweight Prompt Injection (arXiv 2504.16125)", "o": "Academic paper by Xiangyu Chang, Guang Dai, Hao Di, Haishan Ye (from the arXiv API)", "dt": "2025-04-20 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2504.16125", "n": 531, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0522", "t": "Securing Agentic AI: A Comprehensive Threat Model and Mitigation Framework for Generative AI Agents (arXiv 2504.19956)", "o": "Academic paper by Vineeth Sai Narajala, Om Narayan (from the arXiv API)", "dt": "2025-04-28 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2504.19956", "n": 532, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0523", "t": "The Hidden Risks of LLM-Generated Web Application Code: A Security-Centric Evaluation of Code Generation Capabilities in Large Language Models (arXiv 2504.20612)", "o": "Academic paper by Swaroop Dora, Deven Lunkad, Naziya Aslam, S. Venkatesan, Sandeep Kumar Shukla (from the arXiv API)", "dt": "2025-04-29 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2504.20612", "n": 533, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0524", "t": "SecRepoBench: Benchmarking Code Agents for Secure Code Completion in Real-World Repositories (arXiv 2504.21205)", "o": "Academic paper by Chihao Shen, Connor Dilgren, Purva Chiniya, Luke Griffith, Yu Ding, Yizheng Chen (from the arXiv API)", "dt": "2025-04-29 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2504.21205", "n": 534, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0525", "t": "Open Challenges in Multi-Agent Security: Towards Secure Systems of Interacting AI Agents (arXiv 2505.02077)", "o": "Academic paper by Christian Schroeder de Witt, Klaudia Krawiecka, Igor Krawczuk, Ben Hagag, William L. Anderson, Peter Belcak, and 18 others (from the arXiv API)", "dt": "2025-05-04 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2505.02077", "n": 535, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0526", "t": "System Prompt Poisoning: Persistent Attacks on Large Language Models Beyond User Injection (arXiv 2505.06493)", "o": "Academic paper by Zongze Li, Jiawei Guo, Haipeng Cai (from the arXiv API)", "dt": "2025-05-10 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2505.06493", "n": 536, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0527", "t": "AudioJailbreak: Jailbreak Attacks against End-to-End Large Audio-Language Models (arXiv 2505.14103)", "o": "Academic paper by Guangke Chen, Fu Song, Zhe Zhao, Xiaojun Jia, Yang Liu, Yanchen Qiao, and 4 others (from the arXiv API)", "dt": "2025-05-20 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2505.14103", "n": 537, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0528", "t": "BountyBench: Dollar Impact of AI Agent Attackers and Defenders on Real-World Cybersecurity Systems (arXiv 2505.15216)", "o": "Academic paper by Andy K. Zhang, Joey Ji, Celeste Menders, Riya Dulepet, Thomas Qin, Ron Y. Wang, and 28 others (from the arXiv API)", "dt": "2025-05-21 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2505.15216", "n": 538, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0529", "t": "PARASITE: Conditional System Prompt Poisoning to Hijack LLMs (arXiv 2505.16888)", "o": "Academic paper by Viet Pham, Thai Le (from the arXiv API)", "dt": "2025-05-22 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2505.16888", "n": 539, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0530", "t": "OrgAccess: A Benchmark for Role Based Access Control in Organization Scale LLMs (arXiv 2505.19165)", "o": "Academic paper by Debdeep Sanyal, Umakanta Maharana, Yash Sinha, Hong Ming Tan, Shirish Karande, Mohan Kankanhalli, and 1 others (from the arXiv API)", "dt": "2025-05-25 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2505.19165", "n": 540, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0531", "t": "Mind the Gap: A Practical Attack on GGUF Quantization (arXiv 2505.23786)", "o": "Academic paper by Kazuki Egashira, Robin Staab, Mark Vero, Jingxuan He, Martin Vechev (from the arXiv API)", "dt": "2025-05-24 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2505.23786", "n": 541, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0532", "t": "System Prompt Extraction Attacks and Defenses in Large Language Models (arXiv 2505.23817)", "o": "Academic paper by Badhan Chandra Das, M. Hadi Amini, Yanzhao Wu (from the arXiv API)", "dt": "2025-05-27 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2505.23817", "n": 542, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0533", "t": "Seven Security Challenges in Cross-domain Multi-agent LLM Systems (arXiv 2505.23847)", "o": "Academic paper by Ronny Ko, Jiseong Jeong, Shuyuan Zheng, Chuan Xiao, Tae-Wan Kim, Makoto Onizuka, and 1 others (from the arXiv API)", "dt": "2025-05-28 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2505.23847", "n": 543, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0534", "t": "CyberGym: Evaluating AI Agents' Real-World Cybersecurity Capabilities at Scale (arXiv 2506.02548)", "o": "Academic paper by Zhun Wang, Tianneng Shi, Jingxuan He, Matthew Cai, Jialin Zhang, Dawn Song (from the arXiv API)", "dt": "2025-06-03 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2506.02548", "n": 544, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0535", "t": "Design Patterns for Securing LLM Agents against Prompt Injections (arXiv 2506.08837)", "o": "Academic paper by Luca Beurer-Kellner, Beat Buesser, Ana-Maria Creţu, Edoardo Debenedetti, Daniel Dobos, Daniel Fabian, and 8 others (from the arXiv API)", "dt": "2025-06-10 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2506.08837", "n": 545, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0536", "t": "SoK: Evaluating Jailbreak Guardrails for Large Language Models (arXiv 2506.10597)", "o": "Academic paper by Xunguang Wang, Zhenlan Ji, Wenxuan Wang, Zongjie Li, Daoyuan Wu, Shuai Wang (from the arXiv API)", "dt": "2025-06-12 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2506.10597", "n": 546, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0537", "t": "AIRTBench: Measuring Autonomous AI Red Teaming Capabilities in Language Models (arXiv 2506.14682)", "o": "Academic paper by Ads Dawson, Rob Mulla, Nick Landers, Shane Caldwell (from the arXiv API)", "dt": "2025-06-17 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2506.14682", "n": 547, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0538", "t": "OS-Harm: A Benchmark for Measuring Safety of Computer Use Agents (arXiv 2506.14866)", "o": "Academic paper by Thomas Kuntz, Agatha Duzan, Hao Zhao, Francesco Croce, Zico Kolter, Nicolas Flammarion, and 1 others (from the arXiv API)", "dt": "2025-06-17 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2506.14866", "n": 548, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0539", "t": "MAGPIE: A dataset for Multi-AGent contextual PrIvacy Evaluation (arXiv 2506.20737)", "o": "Academic paper by Gurusha Juneja, Alon Albalak, Wenyue Hua, William Yang Wang (from the arXiv API)", "dt": "2025-06-25 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2506.20737", "n": 549, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0540", "t": "Guiding AI to Fix Its Own Flaws: An Empirical Study on LLM-Driven Secure Code Generation (arXiv 2506.23034)", "o": "Academic paper by Hao Yan, Swapneel Suhas Vaidya, Xiaokuan Zhang, Ziyu Yao (from the arXiv API)", "dt": "2025-06-28 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2506.23034", "n": 550, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0541", "t": "From Prompt Injections to Protocol Exploits: Threats in LLM-Powered AI Agents Workflows (arXiv 2506.23260)", "o": "Academic paper by Mohamed Amine Ferrag, Norbert Tihanyi, Djallel Hamouda, Leandros Maglaras, Abderrahmane Lakas, Merouane Debbah (from the arXiv API)", "dt": "2025-06-29 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2506.23260", "n": 551, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0542", "t": "STACK: Adversarial Attacks on LLM Safeguard Pipelines (arXiv 2506.24068)", "o": "Academic paper by Ian R. McKenzie, Oskar J. Hollinsworth, Tom Tseng, Xander Davies, Stephen Casper, Aaron D. Tucker, and 2 others (from the arXiv API)", "dt": "2025-06-30 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2506.24068", "n": 552, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0543", "t": "Meta-SecAlign: Training LLMs against Prompt Injection for Robust Agents (arXiv 2507.02735)", "o": "Academic paper by Sizhe Chen, Arman Zharmagambetov, David Wagner, Chuan Guo (from the arXiv API)", "dt": "2025-07-03 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2507.02735", "n": 553, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0544", "t": "OpenAgentSafety: A Comprehensive Framework for Evaluating Real-World AI Agent Safety (arXiv 2507.06134)", "o": "Academic paper by Sanidhya Vijayvargiya, Aditya Bharat Soni, Xuhui Zhou, Zora Zhiruo Wang, Nouha Dziri, Graham Neubig, and 1 others (from the arXiv API)", "dt": "2025-07-08 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2507.06134", "n": 554, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0545", "t": "The Dark Side of LLMs: Agent-based Attack Vectors for System-level Compromise (arXiv 2507.06850)", "o": "Academic paper by Matteo Lupinacci, Francesco Aurelio Pironti, Francesco Blefari, Francesco Romeo, Luigi Arena, Angelo Furfaro (from the arXiv API)", "dt": "2025-07-09 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2507.06850", "n": 555, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0546", "t": "Manipulating LLM Web Agents with Indirect Prompt Injection Attack via HTML Accessibility Tree (arXiv 2507.14799)", "o": "Academic paper by Sam Johnson, Viet Pham, Thai Le (from the arXiv API)", "dt": "2025-07-20 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2507.14799", "n": 556, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0547", "t": "PromptArmor: Simple yet Effective Prompt Injection Defenses (arXiv 2507.15219)", "o": "Academic paper by Tianneng Shi, Kaijie Zhu, Zhun Wang, Yuqi Jia, Will Cai, Weida Liang, and 10 others (from the arXiv API)", "dt": "2025-07-21 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2507.15219", "n": 557, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0548", "t": "Depth Gives a False Sense of Privacy: LLM Internal States Inversion (arXiv 2507.16372)", "o": "Academic paper by Tian Dong, Yan Meng, Shaofeng Li, Guoxing Chen, Zhen Liu, Haojin Zhu (from the arXiv API)", "dt": "2025-07-22 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2507.16372", "n": 558, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0549", "t": "Resource Consumption Red-Teaming for Large Vision-Language Models (arXiv 2507.18053)", "o": "Academic paper by Haoran Gao, Yuanhe Zhang, Zhenhong Zhou, Lei Jiang, Fanyu Meng, Yujia Xiao, and 4 others (from the arXiv API)", "dt": "2025-07-24 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2507.18053", "n": 559, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0550", "t": "ProbGuard: Proactive Runtime Monitoring for LLM Agent Safety via Probabilistic Prediction (arXiv 2508.00500)", "o": "Academic paper by Haoyu Wang, Christopher M. Poskitt, Jiali Wei, Jun Sun (from the arXiv API)", "dt": "2025-08-01 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2508.00500", "n": 560, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0551", "t": "Exploring Cross-Stage Adversarial Transferability in Class-Incremental Continual Learning (arXiv 2508.08920)", "o": "Academic paper by Jungwoo Kim, Jong-Seok Lee (from the arXiv API)", "dt": "2025-08-12 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2508.08920", "n": 561, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0552", "t": "MCPSecBench: A Systematic Security Benchmark and Playground for Testing Model Context Protocols (arXiv 2508.13220)", "o": "Academic paper by Yixuan Yang, Cuifeng Gao, Daoyuan Wu, Yufan Chen, Yingjiu Li, Shuai Wang (from the arXiv API)", "dt": "2025-08-17 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2508.13220", "n": 562, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0553", "t": "EchoLeak: The First Real-World Zero-Click Prompt Injection Exploit in a Production LLM System (arXiv 2509.10540)", "o": "Academic paper by Pavan Reddy, Aditya Sanjay Gujral (from the arXiv API)", "dt": "2025-09-06 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2509.10540", "n": 563, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0554", "t": "RAG Security and Privacy: Formalizing the Threat Model and Attack Surface (arXiv 2509.20324)", "o": "Academic paper by Atousa Arzanipour, Rouzbeh Behnia, Reza Ebrahimi, Kaushik Dutta (from the arXiv API)", "dt": "2025-09-24 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2509.20324", "n": 564, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0555", "t": "You Can't Steal Nothing: Mitigating Prompt Leakages in LLMs via System Vectors (arXiv 2509.21884)", "o": "Academic paper by Bochuan Cao, Changjiang Li, Yuanpu Cao, Yameng Ge, Ting Wang, Jinghui Chen (from the arXiv API)", "dt": "2025-09-26 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2509.21884", "n": 565, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0556", "t": "Breaking the Code: Security Assessment of AI Code Agents Through Systematic Jailbreaking Attacks (arXiv 2510.01359)", "o": "Academic paper by Shoumik Saha, Jifan Chen, Sam Mayers, Sanjay Krishna Gouda, Zijian Wang, Varun Kumar (from the arXiv API)", "dt": "2025-10-01 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2510.01359", "n": 566, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0557", "t": "Poisoning Attacks on LLMs Require a Near-constant Number of Poison Samples (arXiv 2510.07192)", "o": "Academic paper by Alexandra Souly, Javier Rando, Ed Chapman, Xander Davies, Burak Hasircioglu, Ezzeldin Shereen, and 7 others (from the arXiv API)", "dt": "2025-10-08 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2510.07192", "n": 567, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0558", "t": "The Attacker Moves Second: Stronger Adaptive Attacks Bypass Defenses Against Llm Jailbreaks and Prompt Injections (arXiv 2510.09023)", "o": "Academic paper by Milad Nasr, Nicholas Carlini, Chawin Sitawarin, Sander V. Schulhoff, Jamie Hayes, Michael Ilie, and 8 others (from the arXiv API)", "dt": "2025-10-10 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2510.09023", "n": 568, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0559", "t": "Whisper Leak: a side-channel attack on Large Language Models (arXiv 2511.03675)", "o": "Academic paper by Geoff McDonald, Jonathan Bar Or (from the arXiv API)", "dt": "2025-11-05 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2511.03675", "n": 569, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0560", "t": "Secure Code Generation at Scale with Reflexion (arXiv 2511.03898)", "o": "Academic paper by Arup Datta, Ahmed Aljohani, Hyunsook Do (from the arXiv API)", "dt": "2025-11-05 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2511.03898", "n": 570, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0561", "t": "Replicating TEMPEST at Scale: Multi-Turn Adversarial Attacks Against Trillion-Parameter Frontier Models (arXiv 2512.07059)", "o": "Academic paper by Richard Young (from the arXiv API)", "dt": "2025-12-08 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2512.07059", "n": 571, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0562", "t": "ThinkTrap: Denial-of-Service Attacks against Black-box LLM Services via Infinite Thinking (arXiv 2512.07086)", "o": "Academic paper by Yunzhe Li, Jianan Wang, Hongzi Zhu, James Lin, Shan Chang, Minyi Guo (from the arXiv API)", "dt": "2025-12-08 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2512.07086", "n": 572, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0563", "t": "MIRAGE: Misleading Retrieval-Augmented Generation via Black-box and Query-agnostic Poisoning Attacks (arXiv 2512.08289)", "o": "Academic paper by Tailun Chen, Yu He, Yan Wang, Shuo Shao, Haolun Zheng, Zhihao Liu, and 6 others (from the arXiv API)", "dt": "2025-12-09 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2512.08289", "n": 573, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0564", "t": "MCP-SafetyBench: A Benchmark for Safety Evaluation of Large Language Models with Real-World MCP Servers (arXiv 2512.15163)", "o": "Academic paper by Xuanjun Zong, Zhiqi Shen, Lei Wang, Yunshi Lan, Chao Yang (from the arXiv API)", "dt": "2025-12-17 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2512.15163", "n": 574, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0565", "t": "Constitutional Classifiers++: Efficient Production-Grade Defenses against Universal Jailbreaks (arXiv 2601.04603)", "o": "Academic paper by Hoagy Cunningham, Jerry Wei, Zihan Wang, Andrew Persic, Alwin Peng, Jordan Abderrachid, and 23 others (from the arXiv API)", "dt": "2026-01-08 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2601.04603", "n": 575, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0566", "t": "The Promptware Kill Chain: How Prompt Injections Gradually Evolved Into a Multistep Malware Delivery Mechanism (arXiv 2601.09625)", "o": "Academic paper by Oleg Brodt, Elad Feldman, Bruce Schneier, Ben Nassi (from the arXiv API)", "dt": "2026-01-14 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2601.09625", "n": 576, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0567", "t": "Just Ask: Curious Code Agents Reveal System Prompts in Frontier LLMs (arXiv 2601.21233)", "o": "Academic paper by Xiang Zheng, Yutao Wu, Hanxun Huang, Yige Li, Xingjun Ma, Bo Li, and 2 others (from the arXiv API)", "dt": "2026-01-29 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2601.21233", "n": 577, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0568", "t": "From Similarity to Vulnerability: Key Collision Attack on LLM Semantic Caching (arXiv 2601.23088)", "o": "Academic paper by Zhixiang Zhang, Zesen Liu, Yuchong Xie, Quanfeng Huang, Dongdong She (from the arXiv API)", "dt": "2026-01-30 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2601.23088", "n": 578, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0569", "t": "Zero2Text: Zero-Training Cross-Domain Inversion Attacks on Textual Embeddings (arXiv 2602.01757)", "o": "Academic paper by Doohyun Kim, Donghwa Kang, Kyungjae Lee, Hyeongboo Baek, Brent Byunghoon Kang (from the arXiv API)", "dt": "2026-02-02 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2602.01757", "n": 579, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0570", "t": "Inference-Time Backdoors via Chat Templates: From LLM Supply Chains to Agentic System Compromise (arXiv 2602.04653)", "o": "Academic paper by Ariel Fogel, Omer Hofman, Eilon Cohen, Roman Vainshtein (from the arXiv API)", "dt": "2026-02-04 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2602.04653", "n": 580, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0571", "t": "Clawdrain: Exploiting Tool-Calling Chains for Stealthy Token Exhaustion in OpenClaw Agents (arXiv 2603.00902)", "o": "Academic paper by Ben Dong, Zhonghao Guo, Tianyi Li, Qian Wang (from the arXiv API)", "dt": "2026-03-01 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2603.00902", "n": 581, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0572", "t": "Towards Privacy-Preserving LLM Inference via Covariant Obfuscation (Technical Report) (arXiv 2603.01499)", "o": "Academic paper by Yu Lin, Qizhi Zhang, Wenqiang Ruan, Daode Zhang, Jue Hong, Ye Wu, and 3 others (from the arXiv API)", "dt": "2026-03-02 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2603.01499", "n": 582, "dec": "include", "x": "metadata", "k": "research", "ev": "possibly_deployed", "ax": 1 }, { "i": "WP-0573", "t": "How Vulnerable Are AI Agents to Indirect Prompt Injections? Insights from a Large-Scale Public Competition (arXiv 2603.15714)", "o": "Academic paper by Mateusz Dziemian, Maxwell Lin, Xiaohan Fu, Micha Nowak, Nick Winter, Eliot Jones, and 25 others (from the arXiv API)", "dt": "2026-03-16 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2603.15714", "n": 583, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0574", "t": "SecureForge: Finding and Preventing Vulnerabilities in LLM-Generated Code via Prompt Optimization (arXiv 2605.08382)", "o": "Academic paper by Houjun Liu, Lisa Einstein, John Yang, Joachim Baumann, Duncan Eddy, Christopher D. Manning, and 2 others (from the arXiv API)", "dt": "2026-05-08 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2605.08382", "n": 584, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0575", "t": "ExploitGym: Can AI Agents Turn Security Vulnerabilities into Real Attacks? (arXiv 2605.11086)", "o": "Academic paper by Zhun Wang, Nico Schiller, Hongwei Li, Srijiith Sesha Narayana, Milad Nasr, Nicholas Carlini, and 10 others (from the arXiv API)", "dt": "2026-05-11 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2605.11086", "n": 585, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0576", "t": "Measuring Security Without Fooling Ourselves: Why Benchmarking Agents Is Hard (arXiv 2605.22568)", "o": "Academic paper by Sahar Abdelnabi, Chris Hicks, Konrad Rieck, Ahmad-Reza Sadeghi (from the arXiv API)", "dt": "2026-05-21 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2605.22568", "n": 586, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0577", "t": "CyberGym-E2E: Scalable Real-World Benchmark for AI Agents' End-to-End Cybersecurity Capabilities (arXiv 2606.04460)", "o": "Academic paper by Tianneng Shi, Robin Rheem, Dongwei Jiang, Mona Wang, Francisco De La Riega, Zhun Wang, and 10 others (from the arXiv API)", "dt": "2026-06-03 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2606.04460", "n": 587, "dec": "include", "x": "metadata", "k": "research", "ev": "tested", "ax": 1 }, { "i": "WP-0578", "t": "Every Model Cheats: Prompt-Level Mitigation of Cheating on Offensive Cyber Tasks (arXiv 2607.21763)", "o": "Academic paper by Michael Kouremetis, Ads Dawson, Raja Sekhar Rao Dheekonda, Brian Greunke (from the arXiv API)", "dt": "2026-07-23 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2607.21763", "n": 588, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 }, { "i": "WP-0579", "t": "Contextual Agent Security: A Policy for Every Purpose (arXiv 2501.17070)", "o": "Academic paper by Lillian Tsai, Eugene Bagdasarian (from the arXiv API)", "dt": "2025-01-28 (arXiv v1, from the arXiv API)", "u": "https://arxiv.org/abs/2501.17070", "n": 589, "dec": "include", "x": "metadata", "k": "research", "ev": "theoretical", "ax": 1 } ] }