_version_ 1866911877827657728
author Li, Nathaniel
Pan, Alexander
Gopal, Anjali
Yue, Summer
Berrios, Daniel
Gatti, Alice
Li, Justin D.
Dombrowski, Ann-Kathrin
Goel, Shashwat
Phan, Long
Mukobi, Gabriel
Helm-Burger, Nathan
Lababidi, Rassin
Justen, Lennart
Liu, Andrew B.
Chen, Michael
Barrass, Isabelle
Zhang, Oliver
Zhu, Xiaoyuan
Tamirisa, Rishub
Bharathi, Bhrugu
Khoja, Adam
Zhao, Zhenqi
Herbert-Voss, Ariel
Breuer, Cort B.
Marks, Samuel
Patel, Oam
Zou, Andy
Mazeika, Mantas
Wang, Zifan
Oswal, Palash
Lin, Weiran
Hunt, Adam A.
Tienken-Harder, Justin
Shih, Kevin Y.
Talley, Kemper
Guan, John
Kaplan, Russell
Steneker, Ian
Campbell, David
Jokubaitis, Brad
Levinson, Alex
Wang, Jean
Qian, William
Karmakar, Kallol Krishna
Basart, Steven
Fitz, Stephen
Levine, Mindy
Kumaraguru, Ponnurangam
Tupakula, Uday
Varadharajan, Vijay
Wang, Ruoyu
Shoshitaishvili, Yan
Ba, Jimmy
Esvelt, Kevin M.
Wang, Alexandr
Hendrycks, Dan
author_facet Li, Nathaniel
Pan, Alexander
Gopal, Anjali
Yue, Summer
Berrios, Daniel
Gatti, Alice
Li, Justin D.
Dombrowski, Ann-Kathrin
Goel, Shashwat
Phan, Long
Mukobi, Gabriel
Helm-Burger, Nathan
Lababidi, Rassin
Justen, Lennart
Liu, Andrew B.
Chen, Michael
Barrass, Isabelle
Zhang, Oliver
Zhu, Xiaoyuan
Tamirisa, Rishub
Bharathi, Bhrugu
Khoja, Adam
Zhao, Zhenqi
Herbert-Voss, Ariel
Breuer, Cort B.
Marks, Samuel
Patel, Oam
Zou, Andy
Mazeika, Mantas
Wang, Zifan
Oswal, Palash
Lin, Weiran
Hunt, Adam A.
Tienken-Harder, Justin
Shih, Kevin Y.
Talley, Kemper
Guan, John
Kaplan, Russell
Steneker, Ian
Campbell, David
Jokubaitis, Brad
Levinson, Alex
Wang, Jean
Qian, William
Karmakar, Kallol Krishna
Basart, Steven
Fitz, Stephen
Levine, Mindy
Kumaraguru, Ponnurangam
Tupakula, Uday
Varadharajan, Vijay
Wang, Ruoyu
Shoshitaishvili, Yan
Ba, Jimmy
Esvelt, Kevin M.
Wang, Alexandr
Hendrycks, Dan
contents 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
format Preprint
id arxiv_https___arxiv_org_abs_2403_03218
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearning
Li, Nathaniel
Pan, Alexander
Gopal, Anjali
Yue, Summer
Berrios, Daniel
Gatti, Alice
Li, Justin D.
Dombrowski, Ann-Kathrin
Goel, Shashwat
Phan, Long
Mukobi, Gabriel
Helm-Burger, Nathan
Lababidi, Rassin
Justen, Lennart
Liu, Andrew B.
Chen, Michael
Barrass, Isabelle
Zhang, Oliver
Zhu, Xiaoyuan
Tamirisa, Rishub
Bharathi, Bhrugu
Khoja, Adam
Zhao, Zhenqi
Herbert-Voss, Ariel
Breuer, Cort B.
Marks, Samuel
Patel, Oam
Zou, Andy
Mazeika, Mantas
Wang, Zifan
Oswal, Palash
Lin, Weiran
Hunt, Adam A.
Tienken-Harder, Justin
Shih, Kevin Y.
Talley, Kemper
Guan, John
Kaplan, Russell
Steneker, Ian
Campbell, David
Jokubaitis, Brad
Levinson, Alex
Wang, Jean
Qian, William
Karmakar, Kallol Krishna
Basart, Steven
Fitz, Stephen
Levine, Mindy
Kumaraguru, Ponnurangam
Tupakula, Uday
Varadharajan, Vijay
Wang, Ruoyu
Shoshitaishvili, Yan
Ba, Jimmy
Esvelt, Kevin M.
Wang, Alexandr
Hendrycks, Dan
Machine Learning
Artificial Intelligence
Computation and Language
Computers and Society
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
title The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearning
topic Machine Learning
Artificial Intelligence
Computation and Language
Computers and Society
url https://arxiv.org/abs/2403.03218