作者位次 #4/7ICML 2026Spotlight· Social Aspects / Security
From Poisoned to Aware: Fostering Backdoor Self-Awareness in LLMs
Guangyu Shen · Siyuan Cheng · Xiangzhe Xu · Yuan Zhou · Hanxi Guo · Zhuo Zhang · Xiangyu Zhang
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。
Guangyu Shen · Siyuan Cheng · Xiangzhe Xu · Yuan Zhou · Hanxi Guo · Zhuo Zhang · Xiangyu Zhang
Yutong Gao · Qinglin Meng · Yuan Zhou · Liangming Pan
While Large Language Models (LLMs) have achieved strong performance across many NLP tasks, their opaque internal mechanisms hinder trustworthiness and safe deployment. Existing surveys in explainable AI largely focus on post-hoc explanation methods that interpret trained models through external a...