作者位次 #2/3ICLR 2026Poster
Grokking in LLM Pretraining? Monitor Memorization-to-Generalization without Test
Ziyue Li · Chenrui Fan · Tianyi Zhou
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。
Ziyue Li · Chenrui Fan · Tianyi Zhou
Ming Li · Chenrui Fan · Yize Cheng · Soheil Feizi · Tianyi Zhou
Large language models increasingly expose reasoning traces, yet their underlying cognitive structure and steps remain difficult to identify and analyze beyond surface-level statistics. We adopt Schoenfeld’s Episode Theory as an inductive, intermediate-scale lens and introduce ThinkARM (Anatomy of...
Yize Cheng · Arshia Soltani Moakhar · Chenrui Fan · Parsa Hosseini · Kazem Faghih · Zahra Sodagar · Wenxiao Wang · Soheil Feizi
Large language model (LLM) agents are increasingly used to interact with and execute tasks in dynamic environments. However, a critical yet overlooked limitation of these agents is that they, by default, assume a stationary context, failing to account for the real-world time elapsed between messa...