Researcher — paper.dinq.me

作者位次 #6/7ICLR 2026Poster· Reinforcement Learning / Deep RL

Reinforcement Mid-Training

Yijun Tian · Shaoyu Chen · Zhichao Xu · Yawei Wang · Jinhe Bi · Peng Han · Wei Wang

强化学习与后训练OpenReview

作者位次 #6/9ACL 2026Poster· Natural Language Processing

If an LLM Were a Character, Would It Know Its Own Story? Evaluating Lifelong Learning in LLM s

Siqi Fan · Xiusheng Huang · Yiqun Yao · Xuezhi Fang · Kang Liu · Peng Han · Shuo Shang · Aixin Sun · Yequan Wang

摘要

Large language models (LLMs) can carry out human-like dialogue, but unlike humans, they are stateless due to the superposition property. However, during multi-turn, multi-agent interactions, LLMs begin to exhibit consistent, character-like behaviors—hinting at a form of emergent lifelong learning...

检索与知识智能体与工具评测与基准OpenReview 项目页

Peng Han

获取完整研究档案

联系方式

2026 顶会论文

Reinforcement Mid-Training

If an LLM Were a Character, Would It Know Its Own Story? Evaluating Lifelong Learning in LLM s

常见合作者