一作ICLR 2026Poster· Deep Learning / Everything Else
The First Impression Problem: Internal Bias Triggers Overthinking in Reasoning Models
Renfei Dang · Zhening Li · Shujian Huang · Jiajun Chen
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。
Renfei Dang · Zhening Li · Shujian Huang · Jiajun Chen
Renfei Dang · Peng Hu · Zhejian Lai · Changjiang Gao · Min Zhang · Shujian Huang
Prior works have shown that fine-tuning on new knowledge can induce factual hallucinations in large language models (LLMs), leading to incorrect outputs when evaluated on previously known information. However, the specific manifestations of such hallucination and its underlying mechanisms remain...