作者位次 #3/6ICLR 2026Poster· Social Aspects / Trustworthy Machine Learning
AdvChain: Adversarial Chain-of-Thought Tuning for Robust Safety Alignment of Large Reasoning Models
Zihao Zhu · Xinyu Wu · Gehan Hu · Siwei Lyu · Ke Xu · Baoyuan Wu
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。
Zihao Zhu · Xinyu Wu · Gehan Hu · Siwei Lyu · Ke Xu · Baoyuan Wu