phenomenon · first seen 4 Oct 2026
safety refusal in language models
Also: refusal behavior in language models
Language models declining harmful requests, which can be localized to a small set of removable internal neurons.
Mentions per week
PapersCodeDiscussion
Earliest sources
- Don't Judge an LLM Only by Its Activations: Discovering Suppressed Safety Features via Counterfactual Activation PotentialPapers · 4 Oct 2026 · Swadesh Swain, Sanghamitra Dutta
- CircuitBreak: Pruning a Sparse Safety Circuit of NeuronsPapers · 12 Oct 2026 · Jincheng Ying, Li Wenlin, Wencheng Yang, Yinhao Xiao
- How Firm Are AI Models' Ethical Boundaries? A Pay-Stub Editing Test of 15 ModelsPapers · 22 Oct 2026 · Simiao Ren, Xingyu Shen, Tommy Duong, Tsang Ng, Ankit Raj
- Forcing a Letter Suppresses Refusal: Mechanistic Evidence from WMDPPapers · 5 Nov 2026 · Stepan Epifantsev, Anna Antipova, Oleg Chernikov, Leon Strelkov, Alexey Dontsov
Latest sources
- Forcing a Letter Suppresses Refusal: Mechanistic Evidence from WMDPPapers · 5 Nov 2026 · Stepan Epifantsev, Anna Antipova, Oleg Chernikov, Leon Strelkov, Alexey Dontsov
- How Firm Are AI Models' Ethical Boundaries? A Pay-Stub Editing Test of 15 ModelsPapers · 22 Oct 2026 · Simiao Ren, Xingyu Shen, Tommy Duong, Tsang Ng, Ankit Raj
- CircuitBreak: Pruning a Sparse Safety Circuit of NeuronsPapers · 12 Oct 2026 · Jincheng Ying, Li Wenlin, Wencheng Yang, Yinhao Xiao
- Don't Judge an LLM Only by Its Activations: Discovering Suppressed Safety Features via Counterfactual Activation PotentialPapers · 4 Oct 2026 · Swadesh Swain, Sanghamitra Dutta