作者位次 #3/6ICML 2026Spotlight· Social Aspects / Accountability, Transparency, and Interpretability
Emergent Analogical Reasoning in Transformers
Gouki Minegishi · Jingyuan Feng · Hiroki Furuta · Takeshi Kojima · Yusuke Iwasawa · Yutaka Matsuo
首次获取该候选人的完整档案及 StartLux 评估共 20 积分；每位用户单独解锁，之后查看免费。主动刷新 20 积分。
实时查找电子邮箱，费用不超过20积分；未找到邮箱不扣费。
Gouki Minegishi · Jingyuan Feng · Hiroki Furuta · Takeshi Kojima · Yusuke Iwasawa · Yutaka Matsuo
Yuta Oshima · Daiki Miyake · Kohsei Matsutani · Yusuke Iwasawa · Masahiro Suzuki · Yutaka Matsuo · Hiroki Furuta
Gouki Minegishi · Hiroki Furuta · Takeshi Kojima · Yusuke Iwasawa · Yutaka Matsuo
Emergent misalignment, where fine-tuning on narrow, non-harmful tasks induces harmful behaviors, poses a key challenge for AI safety in LLMs. Despite growing empirical evidence, its underlying mechanism remains unclear. To uncover the reason behind this phenomenon, we propose a mechanistic accoun...