Haibo Qiu 邱海波I am currently working at Meituan as a Researcher. Before that, I received my PhD degree from the School of Computer Science, University of Sydney (USYD), advised by Prof. Dacheng Tao and co-supervised by Prof. Baosheng Yu. Prior to USYD, I obtained my Bachelor's degree in the Department of Electronic Engineering and Information Science from the University of Science and Technology of China (USTC). My research interests lie in multi-modality learning, with a particular focus on the mllm post-training, unified multimodal understanding and generation, and multimodal reasoning model. |
|
Selected Publications†project leader, *equal contribution |
|
Perceptual-Evidence Anchored Reinforced Learning for Multimodal ReasoningChi Zhang, Haibo Qiu†, Qiming Zhang, Yufei Xu, Zhixiong Zeng, Siqi Yang, Peng Shi, Lin Ma and Jing Zhang arXiv preprint arXiv:2511.18437, 2025 
 |
|
Context Cascade Compression: Exploring the Upper Limits of Text CompressionFanfan Liu and Haibo Qiu arXiv preprint arXiv:2511.15244, 2025
|
|
LongCat-Flash-Omni Technical ReportMeituan LongCat Team Technical Report, 2025
|
|
Metis-HOME: Hybrid Optimized Mixture-of-Experts for Multimodal ReasoningXiaohan Lan, Fanfan Liu, Haibo Qiu†, Siqi Yang, Delian Ruan, Peng Shi and Lin Ma arXiv preprint arXiv:2510.20519, 2025
|
|
DeepSketcher: Internalizing Visual Manipulation for Multimodal ReasoningChi Zhang, Haibo Qiu†, Qiming Zhang, Zhixiong Zeng, Lin Ma and Jing Zhang arXiv preprint arXiv:2509.25866, 2025
|
|
STAGE: Stable and Generalizable GRPO for Autoregressive Image GenerationXiaoxiao Ma, Haibo Qiu†, Guohui Zhang, Jie Huang, Zhixiong Zeng, Siqi Yang, Lin Ma and Feng Zhao arXiv preprint arXiv:2509.25027, 2025
|
|
Metis-RISE: RL Incentivizes and SFT Enhances Multimodal Reasoning Model LearningHaibo Qiu, Xiaohan Lan, Fanfan Liu, Xiaohu Sun, Delian Ruan, Peng Shi and Lin Ma arXiv preprint arXiv:2506.13056, 2025
|
|
Towards Better & Faster Autoregressive Image Generation: From the Perspective of EntropyXiaoxiao Ma, Feng Zhao, Pengyang Ling, Haibo Qiu†, Zhixiang Wei, Hu Yu, Jie Huang, Zhixiong Zeng and Lin Ma Conference on Neural Information Processing Systems (NeurIPS), 2025
|
|
UniToken: Harmonizing Multimodal Understanding and Generation through Unified Visual Encodingyang jiao*, Haibo Qiu*, Zequn Jie, Shaoxiang Chen, Jingjing Chen, Lin Ma, and Yu-Gang Jiang Computer Vision and Pattern Recognition Workshop (CVPRW), 2025
|
|
GFNet: Geometric Flow Network for 3D Point Cloud Semantic SegmentationHaibo Qiu, Baosheng Yu and Dacheng Tao Transactions on Machine Learning Research (TMLR), 2022
|
|
SynFace: Face Recognition with Synthetic DataHaibo Qiu, Baosheng Yu, Dihong Gong, Zhifeng Li, Wei Liu and Dacheng Tao International Conference on Computer Vision (ICCV), 2021
|
|
End2End Occluded Face Recognition by Masking Corrupted FeaturesHaibo Qiu, Dihong Gong, Zhifeng Li, Wei Liu and Dacheng Tao IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2021
|
|
Cross View Fusion for 3D Human Pose EstimationHaibo Qiu, Chunyu Wang, Jingdong Wang, Naiyan Wang and Wenjun Zeng International Conference on Computer Vision (ICCV), 2019
|
|
Learning Basis Representation to Refine 3D Human Pose EstimationsChunyu Wang, Haibo Qiu, Alan L. Yuille and Wenjun Zeng AAAI Conference on Artificial Intelligence (AAAI), 2019
|
Academic Services
|
Experiences |
|
Meituan LongCat Multimodal Foundation GroupMultimodal Researcher, 2024.04 - Present |
|
JD Explore AcademyAdvised by Dr. Baosheng Yu and Prof. Dacheng Tao Research intern, 2021.04 - 2022.04 |
|
Tencent AI LabAdvised by Dr. Dihong Gong, Dr. Zhifeng Li and Dr. Wei Liu Research intern, 2019.05 - 2021.03 |
|
Microsoft Research Asia (MSRA)Advised by Dr. Chunyu Wang and Prof. Wenjun Zeng Research intern, 2017.07 - 2018.12 |
|
Last updated in Jan. 2026 |