Researcher — paper.dinq.me

2026 顶会论文

作者位次 #5/5ICML 2026Poster· Theory / Deep Learning

The Devil is in the Condition Numbers: Why is GLU Better than non-GLU Structure?

Xingyu Lyu · Qianqian Xu · zhiyong yang · Peisong Wen · Qingming Huang

理论与学习动力学OpenReview

作者位次 #7/7ICML 2026Poster

ActiveScope: Actively Seeking and Correcting Perception for MLLMs

Yajing Wang · Chao Bi · Junshu Sun · Shufan Shen · Zhaobo Qi · Shuhui Wang · Qingming Huang

OpenReview

作者位次 #6/7ICML 2026Poster· Applications / Computer Vision

Foresee-to-Ground: From Predictive Temporal Perception to Evidence-Driven Reasoning for Video Temporal Grounding

Zelin Zheng · Xinyan Liu · Ruixin Li · Antoni B. Chan · Guorong Li · Qingming Huang · Laiyun Qing

多模态推理OpenReview

作者位次 #3/4ICML 2026Poster· Deep Learning / Graph Neural Networks

Enhancing LLMs for Graph Tasks via Graph-aware LoRA Generation

Junshu Sun · Wanxing Chang · Qingming Huang · Shuhui Wang

效率与系统OpenReview

作者位次 #6/6ICML 2026Oral

GoodDiffusion: Proactive Copyright Protection for Diffusion Bridge Models via Learnable Sample-specific Signatures

Shixi Qin · zhiyong yang · Shilong Bao · Zitai Wang · Qianqian Xu · Qingming Huang

OpenReview

作者位次 #7/7ICML 2026Poster

Localize and Neutralize: Gradient-Guided Token Suppression Against Visual Prompt Injection Attack

Dongpeng Zhang · Ke Ma · Yangbangyan Jiang · Gaozheng Pei · Longtao Huang · Qianqian Xu · Qingming Huang

安全多模态OpenReview

作者位次 #3/4ICML 2026Poster· Deep Learning / Graph Neural Networks

Adaptive Recurrent Message Passing for Test Time Computing on Graphs

Junshu Sun · Wanxing Chang · Qingming Huang · Shuhui Wang

长上下文与架构OpenReview

作者位次 #2/3ICML 2026Poster· Deep Learning / Large Language Models

When LLMs Develop Languages: Symbolic Communication for Efficient Multi-Agent Reasoning

Zhengqi Pei · Qingming Huang · Shuhui Wang

智能体与工具推理效率与系统OpenReview

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

The Bridge-Garden Dilemma in LLM Distillation: Why Mixing Hard and Soft Labels Works

Guanghui Wang · Kaiwen Kacuila · zhiyong yang · Zitai Wang · Jin-Wen Wu · Longtao Huang · Qianqian Xu · Qingming Huang

强化学习与后训练OpenReview

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

From Static to Dynamic: Exploring Self-supervised Image-to-Video Representation Transfer Learning

Yang Liu · Qianqian Xu · Peisong Wen · Siran Dai · Xilin Zhao · Qingming Huang

多模态可解释性OpenReview 项目页

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

Making Training-Free Diffusion Segmentors Scale with the Generative Power

Benyuan Meng · Qianqian Xu · Zitai Wang · XIAOCHUN CAO · Longtao Huang · Qingming Huang

多模态OpenReview 项目页

作者位次 #7/7CVPR 2026Poster· Computer Vision & Pattern Recognition

Mind the Way You Select Negative Texts: Pursuing the Distance Consistency in OOD Detection with VLMs

Zhikang Xu · Qianqian Xu · Zitai Wang · Cong Hua · Sicong Li · zhiyong yang · Qingming Huang

多模态OpenReview 项目页

作者位次 #7/7CVPR 2026Poster· Computer Vision & Pattern Recognition

BlackMirror: Black-Box Backdoor Detection for Text-to-Image Models via Instruction-Response Deviation

Feiran Li · Qianqian Xu · Shilong Bao · zhiyong yang · Xilin Zhao · XIAOCHUN CAO · Qingming Huang

多模态安全OpenReview 项目页

作者位次 #7/7CVPR 2026Poster· Computer Vision & Pattern Recognition

Guiding Diffusion-based Reconstruction with Contrastive Signals for Balanced Visual Representation

Boyu Han · Qianqian Xu · Shilong Bao · zhiyong yang · Ruochen Cui · Xilin Zhao · Qingming Huang

多模态可解释性OpenReview 项目页

作者位次 #7/7CVPR 2026Poster· Computer Vision & Pattern Recognition

Hidden Dangers of Compositional Generation: Diagnosing Semantic Safety Failures in Text-to-Image Models

Haoming Yang · Ke Ma · Ligong Zhang · Xiaojun Jia · Yingfei Sun · Qianqian Xu · Qingming Huang

多模态安全长上下文与架构OpenReview 项目页

作者位次 #5/5AAAI 2026Poster· Artificial Intelligence

DMGINE: Day-Memory Guided Nighttime Image Enhancement for Dynamic Traffic Scenes

Ruizhou Liu · Zhe Wu · Zimo Liu · Qingfang Zheng · Qingming Huang

多模态检索与知识OpenReview

作者位次 #5/5AAAI 2026Poster· Artificial Intelligence

HiGFA: Hierarchical Guidance for Fine-grained Data Augmentation with Diffusion Models

Zhiguang Lu · Qianqian Xu · Peisong Wen · Siran Dai · Qingming Huang

OpenReview

作者位次 #6/6AAAI 2026Poster· Artificial Intelligence

Quantifying the Potential to Escape Filter Bubbles: A Behavior-Aware Measure via Contrastive Simulation

Difu Feng · Qianqian Xu · Zitai Wang · Cong Hua · zhiyong yang · Qingming Huang

评测与基准社会与人本OpenReview

作者位次 #6/6AAAI 2026Poster· Artificial Intelligence

TuckA: Hierarchical Compact Tensor Experts for Efficient Fine-Tuning

Qifeng Lei · zhiyong yang · Qianqian Xu · Cong Hua · Peisong Wen · Qingming Huang

强化学习与后训练效率与系统OpenReview

Qingming Huang

获取完整研究档案

联系方式

2026 顶会论文

The Devil is in the Condition Numbers: Why is GLU Better than non-GLU Structure?

ActiveScope: Actively Seeking and Correcting Perception for MLLMs

Foresee-to-Ground: From Predictive Temporal Perception to Evidence-Driven Reasoning for Video Temporal Grounding

Enhancing LLMs for Graph Tasks via Graph-aware LoRA Generation

GoodDiffusion: Proactive Copyright Protection for Diffusion Bridge Models via Learnable Sample-specific Signatures

Localize and Neutralize: Gradient-Guided Token Suppression Against Visual Prompt Injection Attack

Adaptive Recurrent Message Passing for Test Time Computing on Graphs

When LLMs Develop Languages: Symbolic Communication for Efficient Multi-Agent Reasoning

The Bridge-Garden Dilemma in LLM Distillation: Why Mixing Hard and Soft Labels Works

From Static to Dynamic: Exploring Self-supervised Image-to-Video Representation Transfer Learning

Making Training-Free Diffusion Segmentors Scale with the Generative Power

Mind the Way You Select Negative Texts: Pursuing the Distance Consistency in OOD Detection with VLMs

BlackMirror: Black-Box Backdoor Detection for Text-to-Image Models via Instruction-Response Deviation

Guiding Diffusion-based Reconstruction with Contrastive Signals for Balanced Visual Representation

Hidden Dangers of Compositional Generation: Diagnosing Semantic Safety Failures in Text-to-Image Models

DMGINE: Day-Memory Guided Nighttime Image Enhancement for Dynamic Traffic Scenes

HiGFA: Hierarchical Guidance for Fine-grained Data Augmentation with Diffusion Models

Quantifying the Potential to Escape Filter Bubbles: A Behavior-Aware Measure via Contrastive Simulation

TuckA: Hierarchical Compact Tensor Experts for Efficient Fine-Tuning

常见合作者