{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/the-wmdp-benchmark-measuring-and-reducing","title":"The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearning","arxiv_id":"2403.03218","date":"2024-03-05","proceeding":null,"authors":["Nathaniel Li","Alexander Pan","Anjali Gopal","Summer Yue","Daniel Berrios","Alice Gatti","Justin D. Li","Ann-Kathrin Dombrowski","Shashwat Goel","Long Phan","Gabriel Mukobi","Nathan Helm-Burger","Rassin Lababidi","Lennart Justen","Andrew B. Liu","Michael Chen","Isabelle Barrass","Oliver Zhang","Xiaoyuan Zhu","Rishub Tamirisa","Bhrugu Bharathi","Adam Khoja","Zhenqi Zhao","Ariel Herbert-Voss","Cort B. Breuer","Samuel Marks","Oam Patel","Andy Zou","Mantas Mazeika","Zifan Wang","Palash Oswal","Weiran Lin","Adam A. Hunt","Justin Tienken-Harder","Kevin Y. Shih","Kemper Talley","John Guan","Russell Kaplan","Ian Steneker","David Campbell","Brad Jokubaitis","Alex Levinson","Jean Wang","William Qian","Kallol Krishna Karmakar","Steven Basart","Stephen Fitz","Mindy Levine","Ponnurangam Kumaraguru","Uday Tupakula","Vijay Varadharajan","Ruoyu Wang","Yan Shoshitaishvili","Jimmy Ba","Kevin M. Esvelt","Alexandr Wang","Dan Hendrycks"],"abstract":"The White House Executive Order on Artificial Intelligence highlights the risks of large language models (LLMs) empowering malicious actors in developing biological, cyber, and chemical weapons. To measure these risks of malicious use, government institutions and major AI labs are developing evaluations for hazardous capabilities in LLMs. However, current evaluations are private, preventing further research into mitigating risk. Furthermore, they focus on only a few, highly specific pathways for malicious use. To fill these gaps, we publicly release the Weapons of Mass Destruction Proxy (WMDP) benchmark, a dataset of 3,668 multiple-choice questions that serve as a proxy measurement of hazardous knowledge in biosecurity, cybersecurity, and chemical security. WMDP was developed by a consortium of academics and technical consultants, and was stringently filtered to eliminate sensitive information prior to public release. WMDP serves two roles: first, as an evaluation for hazardous knowledge in LLMs, and second, as a benchmark for unlearning methods to remove such hazardous knowledge. To guide progress on unlearning, we develop RMU, a state-of-the-art unlearning method based on controlling model representations. RMU reduces model performance on WMDP while maintaining general capabilities in areas such as biology and computer science, suggesting that unlearning may be a concrete path towards reducing malicious use from LLMs. We release our benchmark and code publicly at https://wmdp.ai","url_abs":"https://arxiv.org/abs/2403.03218v7","url_pdf":"https://arxiv.org/pdf/2403.03218v7.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"the-wmdp-benchmark-measuring-and-reducing","repo_url":"https://github.com/centerforaisafety/wmdp","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"the-wmdp-benchmark-measuring-and-reducing","repo_url":"https://github.com/locuslab/open-unlearning","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"multiple-choice","task_name":"Multiple-choice"}],"methods":[{"method_slug":"focus","method_name":"Focus"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2403.03218","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2403.03218"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/centerforaisafety/wmdp","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/locuslab/open-unlearning","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":2,"ran_draft_wrong":1},"by_repo_kind":{"listed":{"samples":3,"ran":3,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"5740a4c2fd3b51ed","entry":"forward_with_cache","repo":"centerforaisafety/wmdp","repo_kind":"listed","path":"rmu/utils.py","file_url":"https://github.com/centerforaisafety/wmdp/blob/HEAD/rmu/utils.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"5740a4c2fd3b51ed"}},{"code_sha256_prefix":"db42ee03f6308ed9","entry":"get_params","repo":"centerforaisafety/wmdp","repo_kind":"listed","path":"rmu/utils.py","file_url":"https://github.com/centerforaisafety/wmdp/blob/HEAD/rmu/utils.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"db42ee03f6308ed9"}},{"code_sha256_prefix":"7bdf184a0ba57b39","entry":"load_model","repo":"centerforaisafety/wmdp","repo_kind":"listed","path":"rmu/utils.py","file_url":"https://github.com/centerforaisafety/wmdp/blob/HEAD/rmu/utils.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"7bdf184a0ba57b39"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}