paper.dinq.meAI 与前沿科学顶会 · 学术新星雷达

更新于

返回新星榜

KW

Kun Wang

ICML 2026×6安全多模态可解释性推理强化学习与后训练

新星分贡献者

61

一作贡献0

核心/参与30

方向广度10

合作网络12

0 一作 · 6 篇 · ICML 2026

获取完整研究档案

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。

联系方式

实时查找电子邮箱，费用不超过20积分；未找到邮箱不扣费。

2026 顶会论文

6

作者位次 #7/9ICML 2026Poster· Deep Learning / Large Language Models

SafeSeek: Universal Attribution of Safety Circuits in Language Models

Miao Yu · Siyuan Fu · Moayad Aloqaily · Zhenhong Zhou · Safa Otoum · Xing fan · Kun Wang · Yufei Guo · Qingsong Wen

安全可解释性OpenReview

作者位次 #6/8ICML 2026Poster· Deep Learning / Large Language Models

See First, Reason Later: Mutual Information-Guided Reinforcement Learning for Vision-Language Models

Yin Zhang · Zonghan Wu · Jiaxuan Zhao · Junfeng Fang · Zengxiang Li · Kun Wang · Qingsong Wen · Yilei Shao

推理多模态强化学习与后训练OpenReview

作者位次 #4/8ICML 2026Poster· Deep Learning / Large Language Models

Uncovering Hidden Triggers: Backdoor Attribution in Language Models

Miao Yu · Zhenhong Zhou · Moayad Aloqaily · Kun Wang · Biwei Huang · Stephen Wang · Yueming Jin · Qingsong Wen

安全OpenReview

作者位次 #4/5ICML 2026Poster· Deep Learning / Large Language Models

WET: Mitigating World-Conditioned Knowledge Conflicts via World Entropy Tethering

Zixuan Wang · Yifei He · Zihan Wang · Kun Wang · Chaomeng chen

检索与知识OpenReview

作者位次 #7/8ICML 2026Poster· Deep Learning / Large Language Models

Large Vision-Language Models Get Lost in Attention

Gongli Xi · Ye Tian · Mengyu Yang · Huahui Yi · Liang Lin · Xiaoshuai Hao · Kun Wang · Wendong Wang

多模态OpenReview

作者位次 #5/7ICML 2026Poster· Deep Learning / Large Language Models

LARFT: Closing the Cognition-Action Gap for Length Instruction Following in Large Language Models

Wei Zhang · Lintong Du · yuanhe zhang · Zhenhong Zhou · Kun Wang · Li Sun · Sen Su

常见合作者

ZZZhenhong Zhou×3 QWQingsong Wen×3 MYMiao Yu×2 MAMoayad Aloqaily×2 SFSiyuan Fu SOSafa Otoum XFXing fan YGYufei Guo YZYin Zhang ZWZonghan Wu JZJiaxuan Zhao JFJunfeng Fang