2026 顶会论文 20 作者位次 #10/10 COLM 2026 Poster · Poster Session 1 · Imperial Ballroom · #55
Struct-Searcher: Agentic Structural Thinking Advances Multimodal Deep Information Seeking Fan Zhang · Guibin Zhang · Shengju Qian · Haoxuan Li · Zheng Lian · Hao Wu · Yuan Gao · Xinyu Geng · Xin WANG · Pheng-Ann Heng
We introduce Structural-Searcher, a belief-driven multimodal deep research framework to advance multimodal deep information seeking.
摘要 Deep research agents have attracted increasing attention for their ability to collect large-scale online information to acquire target knowledge, with recent efforts shifting from purely text-based information seeking to multimodal settings. However, existing agentic workflows are largely aligned with Evidence Accumulation Models (EAMs), which linearly aggregate evidence and lack principled mechanisms for handling contradictory information across heterogeneous modalities. Towards this end, we propose Struct-Searcher, a structural agentic workflow grounded in belief revision theory that explicitly maintains an evolving multimodal structural graph throughout the reasoning process, enabling effective conflict-aware multimodal deep information seeking. Extensive experiments across multiple datasets and backbone models demonstrate that Struct-Searcher is: (1) plug-and-play and model-agnostic, yielding an average relative accuracy improvement of 17.2% on BrowseComp-VL across five different backbones. (2) top-performing, consistently outperforming state-of-the-art vision–language models and prior deep research agents, with relative accuracy improvements of 3.7% on MM-BrowseComp, 1.5% on HLE-VL, and 0.7% on BrowseComp-VL over the second-best competing approach.
多模态 检索与知识 推理 Agentic Workflow LLM-based Agents OpenReview 作者位次 #10/10 ICLR 2026 Poster · Applications / Health
KnowGuard: Knowledge-Driven Abstention for Multi-Round Clinical Reasoning Xilin Dang · Kexin Chen · Xiaorui Su · Ayush Noori · Iñaki Arango · Lucas Vittor · LONG XINYI · Yuyang Du · Marinka Zitnik · Pheng-Ann Heng
作者位次 #10/13 ICLR 2026 Poster · Applications / Genetics, Cell Biology, Health, etc
Lost in Tokenization: Context as the Key to Unlocking Biomolecular Understanding in Scientific LLMs Kai Zhuang · Jiawei Zhang · Yumou Liu · Hanqun Cao · Chunbin Gu · Mengdi Liu · Zhangyang Gao · Zitong Jerry Wang · Xuanhe Zhou · Pheng-Ann Heng · Lijun Wu · Conghui He · Cheng Tan
作者位次 #14/15 ICLR 2026 Poster · Applications / Robotics
HybridVLA: Collaborative Diffusion and Autoregression in a Unified Vision-Language-Action Model Jiaming Liu · Hao Chen · Zhuoyang Liu · Pengju An · Renrui Zhang · Chenyang Gu · Xiaoqi Li · Ziyu Guo · Sixiang Chen · Mengzhen Liu · Chengkai Hou · Mengdi Zhao · KC Zhou · Pheng-Ann Heng · Shanghang Zhang
作者位次 #8/8 ICLR 2026 Poster
ProteinAE: Protein Diffusion Autoencoders for Structure Encoding Shaoning Li · Le Zhuo · Yusong Wang · Mingyu Li · Xinheng He · Fandi Wu · Hongsheng Li · Pheng-Ann Heng
作者位次 #8/8 ICLR 2026 Poster · Computer Vision / 3D Rendering & Reconstruction
EgoHandICL: Egocentric 3D Hand Reconstruction with In-Context Learning Binzhu Xie · Shi Qiu · Sicheng Zhang · Yinqiao Wang · Hao Xu · Muzammal Naseer · Chi-Wing Fu · Pheng-Ann Heng
作者位次 #6/9 ICLR 2026 Poster · Computer Vision / Image and Video Generation
reAR: Rethinking Visual Autoregressive Models via Token-wise Consistency Regularization Qiyuan He · Yicong Li · Haotian Ye · Jinghao Wang · Xinyao Liao · Pheng-Ann Heng · Stefano Ermon · James Y Zou · Angela Yao
作者位次 #21/21 ICLR 2026 Poster
MME-Emotion: A Holistic Evaluation Benchmark for Emotional Intelligence in Multimodal Large Language Models Fan Zhang · Zebang Cheng · Chong Deng · Haoxuan Li · Zheng Lian · Qian Chen · Huadai Liu · Wen Wang · YiFan Zhang · Renrui Zhang · Ziyu Guo · Zhihong Zhu · Hao Wu · Haixin Wang · Yefeng Zheng · Xiaojiang Peng · Xian Wu · Kun Wang · Xiangang Li · Jieping Ye · Pheng-Ann Heng
作者位次 #8/12 ICLR 2026 Poster · Applications / Language, Speech and Dialog
Omni-Captioner: Data Pipeline, Models, and Benchmark for Omni Detailed Perception Ziyang Ma · Ruiyang Xu · Zhenghao Xing · Yunfei Chu · Yuxuan Wang · Jinzheng He · Jin Xu · Pheng-Ann Heng · Kai Yu · Junyang Lin · Ensiong Chng · Xie Chen
作者位次 #7/7 CVPR 2026 Poster · Computer Vision & Pattern Recognition
Synergistic Bleeding Region and Point Detection in Laparoscopic Surgical Videos Jialun Pei · Zhangjun Zhou · Diandian Guo · Zhixi Li · Jing Qin · Bo Du · Pheng-Ann Heng
作者位次 #8/9 CVPR 2026 Poster · Computer Vision & Pattern Recognition
Rethinking Intermediate Representation for VLM-based Robot Manipulation Weiliang Tang · Jialin Gao · Jia-Hui Pan · Gang Wang · Li Erran Li · Yun-Hui Liu · Mingyu Ding · Pheng-Ann Heng · Chi-Wing Fu
作者位次 #4/4 CVPR 2026 Poster · Computer Vision & Pattern Recognition
Unlocking Positive Transfer in Incrementally Learning Surgical Instruments: A Self-reflection Hierarchical Prompt Framework Yu Zhu · Kang Li · Zheng Li · Pheng-Ann Heng
作者位次 #9/9 CVPR 2026 Poster · Computer Vision & Pattern Recognition
Thinking-while-Generating: Interleaving Textual Reasoning throughout Visual Generation Ziyu Guo · Renrui Zhang · Hongyu Li · Manyuan Zhang · Xinyan Chen · Sifan Wang · Yan Feng · Peng Pei · Pheng-Ann Heng
作者位次 #12/12 CVPR 2026 Poster · Computer Vision & Pattern Recognition
HiFi-Inpaint: Towards High-Fidelity Reference-Based Inpainting for Generating Detail-Preserving Human-Product Images Yichen Liu · Donghao Zhou · Jie Wang · Xin Gao · Guisheng Liu · Jiatong Li · Quanwei Zhang · Qiang Lyu · Lanqing Guo · Shilei Wen · Weiqiang Wang · Pheng-Ann Heng
作者位次 #8/8 CVPR 2026 Poster · Computer Vision & Pattern Recognition
Benchmarking Endoscopic Surgical Image Restoration and Beyond Jialun Pei · Diandian Guo · Donghui Yang · Zhixi Li · Yuxin Feng · Long Ma · Bo Du · Pheng-Ann Heng
作者位次 #11/12 CVPR 2026 Poster · Computer Vision & Pattern Recognition
From Manuals to Actions: A Unified VLA Model for Chain-of-Thought Manual Generation and Robotic Manipulation Chenyang Gu · Jiaming Liu · Hao Chen · Runzhong Huang · Qingpo Wuwu · Xiaoqi Li · Zhuoyang Liu · Ying Li · Renrui Zhang · Peng Jia · Pheng-Ann Heng · Shanghang Zhang
作者位次 #7/7 KDD 2026 Oral · Knowledge Discovery & Data Mining
CA-DEL: An Open Multi-Target, Multi-Modal Benchmark for Learning from DNA-Encoded Library Screens Mutian He · Hanqun Cao · Cheng Tan · Zijun Gao · Xiaojun Yao · Chunbin Gu · Pheng-Ann Heng
作者位次 #14/15 KDD 2026 Oral · Knowledge Discovery & Data Mining
The Forgetting-Learning Trade-off: Making Reinforcement Learning Work for Protein Language Models Hanqun Cao · Hongrui Zhang · Junde Xu · Zhou Zhang · Lingdong Shen · Minghao Sun · Ge Liu · Jinbo Xu · Wu-Jun Li · Jinren Ni · Cesar de la Fuente-Nunez · Tianfan Fu · Shuting Jin · Pheng-Ann Heng · Fang Wu
作者位次 #8/8 AAAI 2026 Poster · Artificial Intelligence
SurgPub-Video: A Comprehensive Surgical Video Framework for Enhanced Surgical Intelligence in Vision-Language Model Yaoqian Li · Xikai Yang · Dunyuan Xu · Yang Yu · Litao Zhao · Xiaowei Hu · Jinpeng Li · Pheng-Ann Heng
作者位次 #11/11 AAAI 2026 Poster · Artificial Intelligence
IdentityStory: Taming Your Identity-Preserving Generator for Human-Centric Story Generation Donghao Zhou · Jingyu Lin · Guibao Shen · Quande Liu · Jialin Gao · Lihao Liu · Lan Du · Cunjian Chen · Chi-Wing Fu · Xiaowei Hu · Pheng-Ann Heng