Tao Hu · 胡涛HUST · VIBE Lab

RESEARCH OUTPUT

Publications

Research in world models, visual perception, and generative AI.

Preprints

    2026

    1. arXiv
      sa-opsd.png
      Search-Aware On-Policy Self-Distillation with Sibling Alignment for Generative Recommendation
      Rongfeng Guo, Zhuang Zhuang, Ji Dai, Peng Zhao, Shijie Li, Jie Chen, Teng Long, Vincent Tao Hu, and Fei Pan
      Nov 2026
    2. arXiv placeholder
      GramLoop: Training-Free Gram-Gated Replay for Robust Dense Prediction
      Yang Chen, Canyu Shen, Xinzhe Rao, Yuanyi Yan, Yunlu Chen, Meng Tang, Teng Long, and Vincent Tao Hu
      2026
    3. arXiv placeholder
      Training-Free Hidden-State Refinement for Flow-Matching Image Generators
      Yuanyi Yan, Xinzhe Rao, Canyu Shen, Yang Chen, Yunlu Chen, Meng Tang, Teng Long, and Vincent Tao Hu
      2026
    4. arXiv placeholder
      From State to Action: OODA-Tool for Reliable Multi-Turn Tool Use
      Rongfeng Guo, Yinxuan Huang, Yusen Wu, Maoqing Zhong, Yunlu Chen, Meng Tang, Teng Long, and Vincent Tao Hu
      2026
    5. IJCV placeholder
      Diversifying Similar Subjects for Text-to-image Synthesis with Self-Cross Diffusion Guidance and Reward
      Weimin Qiu, Jieke Wang, Zhining Gu, Vincent Tao Hu, and Meng Tang
      International Journal of Computer Vision, 2026
    6. CVPR placeholder
      Guiding Token-Sparse Diffusion Models
      Felix Krause, Stefan Andreas Baumann, Johannes Schusterbauer, Olga Grebenkova, Ming Gui, Vincent Tao Hu, and Björn Ommer
      In CVPR, 2026
    7. ICLR placeholder
      Purrception: Variational Flow Matching for Vector-Quantized Image Generation
      Răzvan-Andrei Matișan, Vincent Tao Hu, Grigory Bartosh, Björn Ommer, Cees G.M. Snoek, Max Welling, Jan-Willem van de Meent, Mohammad Mahdi Derakhshani, and Floor Eijkelboom
      In ICLR, 2026
    8. TPAMI placeholder
      Diffusion Models and Representation Learning: A Survey
      Michael Fuest, Pingchuan Ma, Ming Gui, Johannes Fischer, Vincent Tao Hu, and Björn Ommer
      In T-PAMI, 2026
      The interplay between diffusion models and representation learning

    2025

    1. ICCV placeholder
      TREAD: Token Routing for Efficient Architecture-agnostic Diffusion Training
      Felix Krause, Timy Phan, Ming Gui, Stefan Andreas Baumann, Vincent Tao Hu, and Björn Ommer
      In ICCV, 2025
    2. ICCV placeholder
      Stochastic Interpolants for Revealing Stylistic Flows across the History of Art
      Pingchuan Ma, Ming Gui, Johannes Schusterbauer, Xiaopei Yang, Olga Grebenkova, Vincent Tao Hu, and Björn Ommer
      In ICCV, 2025
    3. CVPR placeholder
      Continuous, Subject-Specific Attribute Control in T2I Models by Identifying Semantic Directions
      Stefan Andreas Baumann, Felix Krause, Michael Neumayr, Nick Stracke, Vincent Tao Hu, and Björn Ommer
      In CVPR, 2025
      Prompt Editing in T2I models
    4. arXiv placeholder
      MaskFlow: Discrete Flows for Flexible and Efficient Long Video Generation
      Michael Fuest, Vincent Tao Hu, and Björn Ommer
      In arXiv, 2025
    5. ICLR placeholder
      ToddlerDiffusion: Flash Interpretable Controllable Diffusion Model
      Eslam Mohamed BAKR, Liangbing Zhao, Vincent Tao Hu, Matthieu Cord, Patrick Perez, and Mohamed Elhoseiny
      In ICLR, 2025
    6. AAAI placeholder
      DepthFM: Fast Monocular Depth Estimation with Flow Matching
      Ming Gui, Johannes S. Fischer, Ulrich Prestel, Pingchuan Ma, Dmytro Kotovenko, Olga Grebenkova, Stefan A. Baumann, Vincent Tao Hu, and Björn Ommer
      In AAAI, 2025
      An exploration of flow matching for blazing fast and zero-shot depth estimation
    7. AAAI placeholder
      Does VLM Classification Benefit from LLM Description Semantics?
      Pingchuan Ma, Lennart Rietdorf, Dmytro Kotovenko, Vincent Tao Hu, and Björn Ommer
      In AAAI, 2025
    8. WACV placeholder
      Distillation of Diffusion Features for Semantic Correspondence
      Frank Fundel, Johannes Schusterbauer, Vincent Tao Hu, and Björn Ommer
      In WACV, 2025

    2024

    1. arXiv
      ./mask.jpg
      [MASK] is All You Need
      Vincent Tao Hu and Björn Ommer
      In arXiv, 2024
    2. arXiv placeholder
      Scaling Image Tokenizers with Grouped Spherical Quantization
      Jiangtao Wang, Zhen Qin, Yifan Zhang, Vincent Tao Hu, Björn Ommer, Rania Briq, and Stefan Kesselheim
      2024
    3. ECCV placeholder
      ZigMa: A DiT-style Zigzag Mamba Diffusion Model
      Vincent Tao Hu, Stefan Andreas Baumann, Ming Gui, Olga Grebenkova, Pingchuan Ma, Johannes Fischer, and Björn Ommer
      In ECCV, 2024
      a DiT-style Mamba-based diffusion models
    4. ECCV placeholder
      Boosting Latent Diffusion with Flow Matching
      Johannes S. Fischer, Ming Gui, Pingchuan Ma, Nick Stracke, Stefan A. Baumann, Vincent Tao Hu, and Björn Ommer
      In ECCV, 2024
      flow matching for super-resolution
    5. arXiv placeholder
      Guided Flow Vision Transformer from Self-Supervised Diffusion Features
      Vincent Tao Hu, Yunlu Chen, Mathilde Caron, Yuki M. Asano, Cees G.M. Snoek, and Björn Ommer
      In arXiv, 2024
    6. arXiv placeholder
      Motion Flow Matching for Human Motion Synthesis and Editing
      Vincent Tao Hu, Wenzhe Yin, Pingchuan Ma, Yunlu Chen, Basura Fernando, Yuki M. Asano, Efstratios Gavves, Pascal Mettes, Björn Ommer, and Cees G.M. Snoek
      In arXiv, 2024
    7. arXiv placeholder
      Training Class-Imbalanced Diffusion Model Via Overlap Optimization
      Divin Yan, Lu Qi, Vincent Tao Hu, Ming-Hsuan Yang, and Meng Tang
      In arXiv, 2024
    8. AAAI
      ./fm.png
      Latent Space Editing in Transformer-based Flow Matching
      Vincent Tao Hu, David W Zhang, Pascal Mettes, Meng Tang, Deli Zhao, and Cees G.M. Snoek
      In AAAI, 2024
    9. EACL
      ./fm-s2s.png
      Flow Matching for Conditional Text Generation in a Few Sampling Steps
      Vincent Tao Hu, Di Wu, Yuki M. Asano, Pascal Mettes, Basura Fernando, Björn Ommer, and Cees G.M. Snoek
      In EACL, 2024
      Flow Matching for text generation

    2023

    1. CVPR
      ./sgdm-why.png
      Self-Guided Diffusion Models
      Tao Hu*, David W Zhang*, Yuki M. Asano, Gertjan J. Burghouts, and Cees G.M. Snoek
      In CVPR, 2023
      A bridge between the community of self-supervised learning and diffusion models. Short version to appear in NeurIPS 2022 Workshop on Score-Based Methods and NeurIPS 2022 Workshop Self-Supervised Learning Theory and Practice.

    2022

    1. WACV
      ./video_retrieval.png
      Self-Supervised Video Representation Learning with Cross-Stream Prototypical Contrasting
      Martine Toering, Ioannis Gatopoulos, Maarten Stol, and Vincent Tao Hu
      In WACV, 2022
      Improve video representation by contrasting prototypical features.

    2020

    1. ECCV
      ./focal.png
      Localizing the Common Action Among a Few Videos
      Pengwan Yang*, Tao Hu*, Pascal Mettes, and Cees G.M. Snoek
      In ECCV, 2020
      Localizing the temporal extent of an action in a long untrimmed video by attention techniques.
    2. ECCV
      ./pointmixup.gif
      Pointmixup: Augmentation for point clouds
      Yunlu Chen*, Tao Hu*, Efstratios Gavves, Thomas Mensink, Pascal Mettes, Pengwan Yang, and Cees G.M. Snoek
      In ECCV, 2020
      A simple augmentation method based on MixUp to boost the performance on related tasks of point cloud.

    2019

    1. AAAI
      ./avatar_amcg.png
      Attention-based Multi-Context Guiding for Few-Shot Semantic Segmentation
      Tao Hu, Pengwan Yang, Chiliang Zhang, Gang Yu, Yadong Mu, and Cees G.M. Snoek
      In AAAI, 2019
      Solve the few-shot segmentation problem by applying attention in multi-scales.
    2. ICCV
      ./avatar_silco.png
      SILCO: Show a Few Images, Localize the Common Object
      Tao Hu, Pascal Mettes, Jia-Hong Huang, and Cees G.M. Snoek
      In ICCV, 2019
      Design a graph network and apply attention on them to solve the problem of common object localization.