Researcher — paper.dinq.me

2026 顶会论文

作者位次 #7/7ICLR 2026Poster· Computer Vision / Vision Models & Multimodal

StreamingVLM: Real-Time Understanding for Infinite Video Streams

Ruyi Xu · Guangxuan Xiao · Yukang Chen · Liuning He · Kelly Peng · Yao Lu · Song Han

多模态OpenReview

作者位次 #7/7ICLR 2026Oral· Deep Learning / Generative Models and Autoencoders

Locality-aware Parallel Decoding for Efficient Autoregressive Image Generation

Zhuoyang Zhang · Luke Huang · Chengyue Wu · Shang Yang · Kelly Peng · Yao Lu · Song Han

多模态效率与系统代码OpenReview

作者位次 #19/20ICLR 2026Oral· Computer Vision / Image and Video Generation

SANA-Video: Efficient Video Generation with Block Linear Diffusion Transformer

Junsong Chen · Yuyang Zhao · Jincheng YU · Ruihang Chu · Junyu Chen · Shuai Yang · Xianbang Wang · Yicheng Pan · Zhou Daquan · Huan Ling · Haozhe Liu · Hongwei Yi · Hao Zhang · Muyang Li · Yukang Chen · Han Cai · Sanja Fidler · Ping Luo · Song Han · Enze Xie

多模态效率与系统OpenReview

作者位次 #13/14ICLR 2026Poster· Reinforcement Learning / Deep RL

QeRL: Beyond Efficiency - Quantization-enhanced Reinforcement Learning for LLMs

Wei Huang · Yi Ge · Shuai Yang · Yicheng Xiao · Huizi Mao · Yujun Lin · Hanrong Ye · Sifei Liu · Ka Chun Cheung · Hongxu (Danny) Yin · Yao Lu · XIAOJUAN QI · Song Han · Yukang Chen

强化学习与后训练效率与系统OpenReview

作者位次 #7/13ICLR 2026Poster· Computer Vision / Vision Models & Multimodal

3D Aware Region Prompted Vision Language Model

An-Chieh Cheng · Yang Fu · Yukang Chen · Zhijian Liu · Xiaolong Li · Subhashree Radhakrishnan · Song Han · Yao Lu · Jan Kautz · Pavlo Molchanov · Hongxu (Danny) Yin · Xiaolong Wang · Sifei Liu

多模态OpenReview

作者位次 #29/32ICLR 2026Poster· Computer Vision / Vision Models & Multimodal

OmniVinci: Enhancing Architecture and Data for Omni-Modal Understanding LLM

Hanrong Ye · Chao-Han Huck Yang · Arushi Goel · Wei Huang · Zhen Wan · Jinchuan Tian · An-Chieh Cheng · Ligeng Zhu · Yuanhang Su · Yuming Lou · Yong-Xiang Lin · Dong Yang · Sreyan Ghosh · Zhijian Liu · Yukang Chen · Ehsan Jahangiri · Ambrish Dantrey · Daguang Xu · Ehsan Hosseini-Asl · Seyed Danial Mohseni Taheri · Vidya Nariyambut Murali · Sifei Liu · Yao Lu · Oluwatobi Olabiyi · Yu-Chiang Frank Wang · Rafael Valle · Bryan Catanzaro · Andrew Tao · Song Han · Jan Kautz · Hongxu (Danny) Yin · Pavlo Molchanov

多模态长上下文与架构OpenReview

作者位次 #3/4ICLR 2026Poster· Deep Learning / Generative Models and Autoencoders

ParoQuant: Pairwise Rotation Quantization for Efficient Reasoning LLM Inference

Yesheng Liang · Haisheng Chen · Song Han · Zhijian Liu

效率与系统推理OpenReview

作者位次 #11/12ICLR 2026Poster

LongLive: Real-time Interactive Long Video Generation

Shuai Yang · Wei Huang · Ruihang Chu · Yicheng Xiao · Yuyang Zhao · Xianbang Wang · Muyang Li · Enze Xie · YINGCONG CHEN · Yao Lu · Song Han · Yukang Chen

多模态OpenReview

作者位次 #8/9ICLR 2026Poster

Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decoding

Chengyue Wu · Hao Zhang · Shuchen Xue · Zhijian Liu · Shizhe Diao · Ligeng Zhu · Ping Luo · Song Han · Enze Xie

效率与系统代码OpenReview

作者位次 #9/10ICLR 2026Poster

Fast-dLLM v2: Efficient Block-Diffusion LLM

Chengyue Wu · Hao Zhang · Shuchen Xue · Shizhe Diao · Yonggan Fu · Zhijian Liu · Pavlo Molchanov · Ping Luo · Song Han · Enze Xie

效率与系统OpenReview

作者位次 #8/15CVPR 2026Poster· Computer Vision & Pattern Recognition

Grounded 3D-Aware Spatial Vision-Language Modeling

An-Chieh Cheng · Yang Fu · Yatai Ji · Ligeng Zhu · Guanqi Zhan · Zhuoyang Zhang · Zhaojing Yang · Song Han · Yao Lu · Pavlo Molchanov · Vidya Nariyambut Murali · Jan Kautz · Xiaolong Wang · Hongxu Yin · Sifei Liu

多模态OpenReview 项目页

作者位次 #13/15CVPR 2026Poster· Computer Vision & Pattern Recognition

DeltaQuant: 4-bit Video Diffusion Models with Spatiotemporal Delta Smoothing

Xingyang Li · Samuel Tesfai · Zhekai Zhang · Haocheng Xi · Shuo Yang · Lvmin Zhang · Yufei Sun · Kelly Peng · Maneesh Agrawala · Ion Stoica · Kurt Keutzer · Jun-Yan Zhu · Song Han · Yujun Lin · Muyang Li

多模态OpenReview 项目页

作者位次 #8/8CVPR 2026Spotlight· Computer Vision & Pattern Recognition

ForeAct: Steering Your VLA with Efficient Visual Foresight Planning

Zhuoyang Zhang · Shang Yang · Qinghao Hu · Luke J. Huang · James Hou · Yufei Sun · Yao Lu · Song Han

多模态推理可解释性OpenReview 项目页

作者位次 #10/10IROS 2026Poster· Efficiency and Safety in Vision-Language-Action Models

VLASH: Real-Time VLAs Via Future-State-Aware Asynchronous Inference

Jiaming Tang · Yufei Sun · Yilong Zhao · Shang Yang · Yujun Lin · Zhuoyang Zhang · James Hou · Yao Lu · Zhijian Liu · Song Han

效率与系统安全多模态AI-Enabled RoboticsMachine Learning for Robot ControlImitation LearningOpenReview

作者位次 #13/15IROS 2026Poster· New Representations for Vision-Language-Action Robots

EgoVLA: Learning Vision-Language-Action Models from Egocentric Human Videos

Ruihan Yang · Qinxi Yu · Wu, Yecheng, Yecheng · Rui Yan · borui li · An-Chieh Cheng · Zou Xueyan · Yunhao Fang · Xuxin Cheng · Ri-Zhao Qiu · Hongxu Yin · Sifei Liu · Song Han · Yao Lu · Xiaolong Wang

多模态可解释性社会与人本Learning from DemonstrationImitation LearningMachine Learning for Robot ControlOpenReview

Song Han

获取完整研究档案

联系方式

2026 顶会论文

StreamingVLM: Real-Time Understanding for Infinite Video Streams

Locality-aware Parallel Decoding for Efficient Autoregressive Image Generation

SANA-Video: Efficient Video Generation with Block Linear Diffusion Transformer

QeRL: Beyond Efficiency - Quantization-enhanced Reinforcement Learning for LLMs

3D Aware Region Prompted Vision Language Model

OmniVinci: Enhancing Architecture and Data for Omni-Modal Understanding LLM

ParoQuant: Pairwise Rotation Quantization for Efficient Reasoning LLM Inference

LongLive: Real-time Interactive Long Video Generation

Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decoding

Fast-dLLM v2: Efficient Block-Diffusion LLM

Grounded 3D-Aware Spatial Vision-Language Modeling

DeltaQuant: 4-bit Video Diffusion Models with Spatiotemporal Delta Smoothing

ForeAct: Steering Your VLA with Efficient Visual Foresight Planning

VLASH: Real-Time VLAs Via Future-State-Aware Asynchronous Inference

EgoVLA: Learning Vision-Language-Action Models from Egocentric Human Videos

常见合作者