👋 About Me

I am Henghui Du, a Master student at Gaoling School of Artificial Intelligence, Renmin University of China, advised by Prof. Di Hu. I am a member of GeWu-Lab. Before that, I received my B.E. degree from Dalian University of Technology in 2023.

My research interests lie in Multimodal Large Language Models (MLLMs), Audio-Visual Scene Understanding and Reasoning. I am dedicated to building unified models that can perceive and reason across multiple modalities including vision, audio, and language.

🔥 News

2026.03 🚀 One paper is accepted by CVPR 2026!
2025.02 🎉 One paper is accepted by CVPR 2025!
2024.07 📝 Two papers are accepted by ACM MM 2024 and TPAMI 2024.
2023.09 🎓 Started my Master's program at Renmin University of China, joining GeWu-Lab.
2023.06 🎓 Graduated from Dalian University of Technology with a B.E. degree.

📝 Publications

* denotes equal contribution. Full list on Google Scholar.

Crab v2

Crab+: A Scalable and Unified Audio-Visual Scene Understanding Model with Explicit Cooperation

Dongchao Cai, Henghui Du, Chaoqun Zhou, Xi Chen, Daguang Guo, Haozhe Zhang, Xinlong Li, Di Hu

arXiv preprint, 2026

APPO

APPO: Attention-guided Perception Policy Optimization for Video Reasoning

Henghui Du, Chaoqun Zhou, Xi Chen, Di Hu

CVPR, 2026

Video Detective

Video Detective: Seek Critical Clues Recurrently to Answer Question from Long Videos

Henghui Du, Chenxu Zhang, Xi Chen, Chaoqun Zhou, Di Hu

arXiv preprint, 2025

Crab

Crab: A Unified Audio-Visual Scene Understanding Model with Explicit Cooperation

Henghui Du, Guangyao Li, Chaoqun Zhou, Chenxu Zhang, Anni Zhao, Di Hu

CVPR, 2025

AVQA

Boosting Audio Visual Question Answering via Key Semantic-Aware Cues

Guangyao Li, Henghui Du, Di Hu

ACM MM, 2024

OFM

On-the-fly Modulation for Balanced Multimodal Learning

Yake Wei, Di Hu, Henghui Du, Ji-Rong Wen

IEEE TPAMI, 2024

AVQA-CoT

AVQA-CoT: When CoT Meets Question Answering in Audio-Visual Scenarios

Guangyao Li, Henghui Du, Di Hu

CVPR Sight and Sound Workshops , 2024.

🎓 Education

Renmin University of China

2023.09 - Present

M.S. in Artificial Intelligence, Gaoling School of Artificial Intelligence

Advisor: Prof. Di Hu

Dalian University of Technology

2019.09 - 2023.06

B.E. degree