Researcher — paper.dinq.me

2026 顶会论文

作者位次 #3/6ICLR 2026Poster· Reinforcement Learning / Online

From Observations to Events: Event-Aware World Models for Reinforcement Learning

Peng Zhao-Han · Shaohui Li · Zhi Li · Shulan Ruan · Yu LIU · You He

强化学习与后训练OpenReview

作者位次 #4/6CVPR 2026Poster· Computer Vision & Pattern Recognition

From Attraction to Equilibrium: Physics-Inspired Semantic Gravitons for Zero-Shot Anomaly Detection

Yuwen Pan · Yuan Wang · Shaohui Li · Zhi Li · Yu LIU · You He

多模态OpenReview 项目页

一作ACL 2026Findings· Findings of ACL

Entropy-Aware Reshaping of Reinforcement Signals for Multi-Answer Reasoning

Zhi Li · Huidan Xu · Zhen Hu · Yali Du · Ying Liu

摘要

Reinforcement learning with verifiable rewards (RLVR) is a standard post-training paradigm for large language models (LLMs), typically relying on group-wise reward and advantage normalization for stability. In set-valued multi-answer tasks, where multiple outputs may be simultaneously correct, th...

强化学习与后训练推理对齐与偏好OpenReview 项目页

作者位次 #2/6IROS 2026Poster· Controlling Marine Vehicles

Formation-Scalable Multi-USV Path Following in Rough Sea States Via Reinforcement Learning with Collective Graph

Zhiqiang Lu · Zhi Li · Di Wu · Shaohui Li · Shiwei Wu · Yu LIU

强化学习与后训练Marine RoboticsAI-Based MethodsOpenReview

Zhi Li

获取完整研究档案

联系方式

2026 顶会论文

From Observations to Events: Event-Aware World Models for Reinforcement Learning

From Attraction to Equilibrium: Physics-Inspired Semantic Gravitons for Zero-Shot Anomaly Detection

Entropy-Aware Reshaping of Reinforcement Signals for Multi-Answer Reasoning

Formation-Scalable Multi-USV Path Following in Rough Sea States Via Reinforcement Learning with Collective Graph

常见合作者