Researcher — paper.dinq.me

2026 顶会论文

作者位次 #4/6ICML 2026Poster· Deep Learning / Large Language Models

Tiny Brains, Giant Impact: Uncovering the Keystone Neurons of LLM with Just a Few Prompts

Xiangtian Ji · Yuxin Chen · Zhengzhou Cai · Xiang Wang · An Zhang · Tat-Seng Chua

可解释性OpenReview

作者位次 #6/8ICML 2026Poster

Contrastive Weak-to-Strong Generalization

Houcheng Jiang · Junfeng Fang · Jiaxin Wu · Tianyu Zhang · Chen Gao · Xiang Wang · Xiangnan He · Yang Deng

理论与学习动力学OpenReview

作者位次 #7/8ICML 2026Poster

Enhancing Multi-Modal LLMs Reasoning via Difficulty-Aware Group Normalization

Jinghan Li · Junfeng Fang · Jinda Lu · Yuan Wang · Xiaoyan Guo · Tianyu Zhang · Xiang Wang · Xiangnan He

推理多模态OpenReview

作者位次 #8/10ICML 2026Poster

Experience Augmented Policy Optimization for LLM Reasoning

Jinda Lu · Kexin Huang · Junkang Wu · Shuo Yang · Jinghan Li · Chiyu Ma · Shaohang Wei · Xiang Wang · Guoyin Wang · Jingren Zhou

推理强化学习与后训练OpenReview

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

Transport and Merge: Cross-Architecture Merging for Large Language Models

Chenhang Cui · Binyun Yang · Fei Shen · Yuxin Chen · Jingnan Zheng · Xiang Wang · An Zhang · Tat-Seng Chua

长上下文与架构OpenReview

作者位次 #7/7ICML 2026Poster

NExT-Guard: Training-Free Streaming Safeguard without Token-Level Labels

Junfeng Fang · Nachuan Chen · Houcheng Jiang · Dan Zhang · Xiangnan He · Tat-Seng Chua · Xiang Wang

OpenReview

作者位次 #2/5ICML 2026Poster· Social Aspects / Safety

Position: Safe AI Should be Resistant and Resilient in an Evolving World

Youbang Sun · Xiang Wang · Jie Fu · Chaochao Lu · Bowen Zhou

安全社会与人本OpenReview

作者位次 #10/12ICML 2026Poster· Social Aspects / Robustness

AgentNoiseBench: Benchmarking Robustness of Tool-Using LLM Agents Under Noisy Condition

Ruipeng Wang · Yuxin Chen · Yukai Wang · Chang Wu · Junfeng Fang · Xiaodong Cai · Qi GU · Hui Su · An Zhang · Xiang Wang · Xunliang Cai · Tat-Seng Chua

智能体与工具评测与基准社会与人本OpenReview

作者位次 #6/7ICML 2026Poster· Social Aspects / Safety

Internalizing Safety Understanding in Large Reasoning Models via Verification

Yi Zhang · Yuxin Chen · Leheng Sheng · Dongcheng Zhang · Chaochao Lu · Xiang Wang · An Zhang

推理安全社会与人本OpenReview

作者位次 #6/9ICML 2026Poster

When to Memorize and When to Stop: Gated Recurrent Memory for Long-Context Reasoning

Leheng Sheng · Yongtao Zhang · Wenchang Ma · Yaorui Shi · Ting Huang · Xiang Wang · An Zhang · Ke Shen · Tat-Seng Chua

长上下文与架构推理检索与知识OpenReview

作者位次 #10/12ICML 2026Poster

Learning to Self-Verify Makes Language Models Better Reasoners

Yuxin Chen · Yu Wang · Yi Zhang · Ziang Ye · Zhengzhou Cai · Yaorui Shi · Qi GU · Hui Su · Xunliang Cai · Xiang Wang · An Zhang · Tat-Seng Chua

推理OpenReview

作者位次 #9/10ICML 2026Poster

MemOCR: Layout-Aware Visual Memory for Efficient Long-Horizon Reasoning

Yaorui Shi · Shugui Liu · Yu Yang · Wenyu Mao · Yuxin Chen · Qi GU · Hui Su · Xunliang Cai · Xiang Wang · An Zhang

推理多模态效率与系统OpenReview

作者位次 #5/6ICML 2026Poster

Mitigating Reward Hacking in LLM-based Recommendation: A Preference Optimization Approach

Heyu Chen · Junkang Wu · Guoqing Hu · Kexin Huang · Xiang Wang · Jiancan Wu

对齐与偏好OpenReview

作者位次 #4/5ICML 2026Poster

Reasoning Can Be Restored by Correcting a Few Decision Tokens

Shen Changshuo · Leheng Sheng · Yuxin Chen · Xiang Wang · An Zhang