作者位次 #4/6ICML 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
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...
Yusuke Watanabe · Yohei Kobashi · Takeshi Kojima · Yusuke Iwasawa · Yasushi Okuno · Yutaka Matsuo
Clinical decisions are often required under incomplete information. Clinical experts must identify whether available information is sufficient for judgment, as both premature conclusions and unnecessary abstention can compromise patient safety. To evaluate this capability of large language models...
Qi Cao · Takeshi Kojima · Andrew Gambardella · Helinyi Peng · Yutaka Matsuo · Yusuke Iwasawa
Large language models (LLMs) demonstrate remarkable performance across diverse tasks, but they often generate responses that appear plausible while being factually incorrect. This problem is compounded by the lack of explicit uncertainty estimates, which makes it difficult for users to judge the...