huangrh9-photo

Runhui Huang (黄润辉)

huangrh9@gmail.com

Hi, I am currently a PhD student at the University of Hong Kong, under the supervision of Prof. Hengshuang Zhao. Previously, I received my Bachelor's degree (2021) and Master's degree (2024) from Sun Yat-sen University, where I was supervised by Prof. Xiaodan Liang.

My research focuses on Unified Multimodal Understanding and Generation, with broad interests in vision-language pre-training and post-training.

More about me: Google Scholar / Github

Research Interests

  • Unified Multimodal Understanding and Generation
  • Reinforcement Learning for Multimodal Models

News

  • [07/2026]We released SpectraReward, a training-free reward model for text-to-image generation.
  • [05/2026]AlphaGRPO was accepted to ICML 2026.
  • [09/2025]I joined ByteDance Seed as a Research Intern.
  • [02/2025]Two papers, HiRes-LLaVA and EMOVA, were accepted to CVPR 2025.
  • [09/2024]I started my Ph.D. at HKU.
  • [09/2024]I joined Huawei Noah's Ark Lab (Hong Kong) as a Research Intern.
  • [07/2024]LayerDiff was accepted to ECCV 2024.

Selected Publications [Google Scholar]

AlphaGRPO training framework
AlphaGRPO: Unlocking Self-Reflective Multimodal Generation in Unified Multimodal Models via Decompositional Verifiable Reward
Runhui Huang, Jie Wu, Rui Yang, Zhe Liu, Hengshuang Zhao
ICML, 2026.
[paper] [project] [code] [models]
ILLUME+ multimodal understanding and image generation examples
ILLUME+: Illuminating Unified MLLM with Dual Visual Tokenization and Diffusion Refinement
Runhui Huang, Chunwei Wang, Junwei Yang, Guansong Lu, Yunlong Yuan, Jianhua Han, Lu Hou, Wei Zhang, Lanqing Hong, Hengshuang Zhao, Hang Xu
CVPR, 2026.
[paper] [project] [code] [demo] [models]
EMOVA model architecture
EMOVA: Empowering Language Models to See, Hear and Speak with Vivid Emotions
Kai Chen*, Yunhao Gou*, Runhui Huang*, Zhili Liu*, Daxin Tan*, Jing Xu, Chao Wang, Yi Zhu, Yihan Zeng.
CVPR, 2025.
[paper] [project] [code]
HiRes-LLaVA model architecture
HiRes-LLaVA: Restoring Fragmentation Input in High-Resolution Large Vision-Language Models
Runhui Huang, Xinpeng Ding, Chunwei Wang, Jianhua Han, Yulong Liu, Hengshuang Zhao, Hang Xu, Lu Hou, Wei Zhang, Xiaodan Liang.
CVPR, 2025.
[paper]
LayerDiff model architecture
LayerDiff: Exploring Text-guided Multi-layered Composable Image Synthesis via Layer-Collaborative Diffusion Model
Runhui Huang, Kaixin Cai, Jianhua Han, Xiaodan Liang, Renjing Pei, Guansong Lu, Songcen Xu, Wei Zhang, Hang Xu
ECCV, 2024.
[paper]
FILIP model architecture
FILIP: Fine-grained Interactive Language-Image Pre-Training
Lewei Yao∗, Runhui Huang∗, Lu Hou∗, Guansong Lu, Minzhe Niu, Hang Xu, Xiaodan Liang, Zhenguo Li, Xin Jiang, Chunjing Xu.
International Conference on Learning Representations (ICLR), 2022.
[paper]

Selected Honors

  • 2023 Academic Innovation Scholarship, Sun Yat-sen University
  • 2022 Outstanding Student Scholarship (First Class), Sun Yat-sen University
  • 2019 Second Prize, National College Student Mathematical Modeling Competition (Guangdong)

Professional Activities

  • Reviewer: CVPR 2023-2026, ICCV 2023/2025, ECCV 2024/2026, ICLR 2025-2026, NeurIPS 2024-2025, ICML 2026 (Gold Reviewer)