Hugging Face Trending Papers

Do LLMs Know Their Vulnerable Scenarios?

Read the original on Hugging Face Trending Papers →

Safety-aligned large language models are trained to refuse harmful requests, yet embedding the same requests in particular scenarios can bypass their safeguards. Existing red-teaming methods empirically identify effective scenarios through observed attack outcomes, but why particular scenarios weaken refusal remains mechanistically unclear.

Summary generated by The Flow from the publisher's feed. The full article lives at Hugging Face Trending Papers.

arXiv AI
Jul 28

Do LLMs Know Their Vulnerable Scenarios?

arXiv:2607. 23496v1 Announce Type: new Abstract: Safety-aligned large language models are trained to refuse harmful requests, yet embedding the same requests in particular scenarios can bypass their safeguards.

By Ziheng Peng, Huiqi Deng, Haoran Jing, Xuankun Rong, Jiahui Han, Xiting Wang, Na Zou, Xia Hu