TensorX
返回文献探索

Paper · arXiv 2510.11288

Emergent Misalignment via In-Context Learning: Narrow in-context examples can produce broadly misaligned LLMs

Nikita Afonin, Nikita Andriyanov, Nikhil Bageshpura, Kyle Liu, Kevin Zhu, Sunishchal Dev, Ashwinee Panda, Alexander Panchenko, Oleg Rogov, Elena Tutubalina, Mikhail Seleznyov

48 upvotesOctober 13, 2025arXiv 预印本
AI 摘要

Emergent misalignment occurs in in-context learning across multiple models and datasets, with misaligned responses increasing with the number of examples provided.

emergent misalignmentin-context learningchain-of-thoughtpersona

Abstract

Recent work has shown that narrow finetuning can produce broadly misaligned LLMs, a phenomenon termed emergent misalignment (EM). While concerning, these findings were limited to finetuning and activation steering, leaving out in-context learning (ICL). We therefore ask: does EM emerge in ICL? We find that it does: across three datasets, three frontier models produce broadly misaligned responses at rates between 2% and 17% given 64 narrow in-context examples, and up to 58% with 256 examples. We also examine mechanisms of EM by eliciting step-by-step reasoning (while leaving in-context examples unchanged). Manual analysis of the resulting chain-of-thought shows that 67.5% of misaligned traces explicitly rationalize harmful outputs by adopting a reckless or dangerous ''persona'', echoing prior results on finetuning-induced EM.

北京市昌平区探索星信息技术及软件开发工作室

京ICP备2026059466号
Emergent Misalignment via In-Context Learning: Narrow in-context examples can produce broadly misaligned LLMs | TensorX