Publications

The year indicates the time when the work is mostly finished.
* indicates equal contribution.

2025

  1. Hunyuan3D Studio: End-to-End AI Pipeline for Game-Ready 3D Asset Generation
    Preprint, arXiv preprint arXiv:2509.12815.
    Biwen Lei, Yang Li, Xinhai Liu, Shuhui Yang, Lixin Xu, Jingwei Huang, Ruining Tang, Haohan Weng, Jian Liu, Jing Xu, Zhen Zhou, Yiling Zhu, and 88 more authors
  2. Hunyuan3D 3.0: System Card
    Preprint, Report.
    Zeqiang Lai, Yunfei Zhao, Haolin Liu, Zibo Zhao, Qingxiang Lin, Zhihao Liang, Bowen Zhang, Xianghui Yang, Yifei Huang, Xin Huang, Li Wang, Xin Yang, and 12 more authors
  3. X-Part: High Fidelity and Structure Coherent Shape Decomposition
    Preprint, arXiv preprint arXiv:2509.08643. [Code]
    Xinhao Yan, Jiachen Xu, Yang Li, Changfeng Ma, Yunhan Yang, Chunshi Wang, Zibo Zhao, Zeqiang Lai, Yunfei Zhao, Zhuo Chen, and Chunchao Guo
  4. Hunyuan3D 2.5: Towards High-Fidelity 3D Assets Generation with Ultimate Details
    Preprint, arXiv preprint arXiv:2506.16504. [Code]
    Zeqiang Lai*, Yunfei Zhao*, Haolin Liu*, Zibo Zhao, Qingxiang Lin, Huiwen Shi, Xianghui Yang, Mingxin Yang*, Shuhui Yang*, Yifei Feng*, Sheng Zhang, Xin Huang, and 14 more authors
  5. Hunyuan3D 2.1: From Images to High-Fidelity 3D Assets with Production-Ready PBR Material
    Preprint, arXiv preprint arXiv:2506.15442. [Code]
    Shuhui Yang, Mingxin Yang, Yifei Feng, Xin Huang, Sheng Zhang, Zebin He, Di Luo, Haolin Liu, Yunfei Zhao, Qingxiang Lin, Zeqiang Lai, Xianghui Yang, and 40 more authors
  6. FlashVDM: Unleashing Vecset Diffusion Model for Fast Shape Generation
    ICCV 2025 Highlight, International Conference on Computer Vision. [Code]
    Zeqiang Lai*, Yunfei Zhao*, Zibo Zhao, Haolin Liu, Fuyun Wang, Huiwen Shi, Xianghui Yang, Qingxiang Lin, Jingwei Huang, Yuhong Liu, Jie Jiang, Chunchao Guo, and 1 more author
  7. Hunyuan3D 2.0: Scaling Diffusion Models for High Resolution Textured 3D Assets Generation
    Preprint, arXiv preprint arXiv:2501.12202. [Website] [Code]
    Zibo Zhao*, Zeqiang Lai*, Qingxiang Lin*, Yunfei Zhao*, Haolin Liu, Shuhui Yang*, Yifei Feng*, Mingxin Yang*, Sheng Zhang*, Xianghui Yang, Huiwen Shi, Sicong Liu, and 59 more authors

2024

  1. MuLan: Adapting Multilingual Diffusion Models for Hundreds of Languages with Negligible Cost
    ICML 2025, International Conference on Machine Learning. [Code]
    Sen Xing*, Muyan Zhong*, Zeqiang Lai*, Liangchen Li, Jiawen Liu, Yaohui Wang, Jifeng Dai, and Wenhai Wang
  2. Flexitex: Enhancing Texture Generation with Visual Guidance
    AAAI 2025, AAAI Conference on Artificial Intelligence. [Website]
    DaDong Jiang, Xianghui Yang, Zibo Zhao, Sheng Zhang, Jiaao Yu, Zeqiang Lai, Shaoxiong Yang, Chunchao Guo, Xiaobo Zhou, and Zhihui Ke
  3. Scaling Mesh Generation via Compressive Tokenization
    CVPR 2025, The IEEE/CVF Conference on Computer Vision and Pattern Recognition. [Website] [Code]
    Haohan Weng, Zibo Zhao, Biwen Lei, Xianghui Yang, Jian Liu, Zeqiang Lai, Zhuo Chen, Yuhong Liu, Jie Jiang, Chunchao Guo, and  others
  4. VisionLLM v2: An End-to-End Generalist Multimodal Large Language Model for Hundreds of Vision-Language Tasks
    NeurIPS 2024, Conference on Neural Information Processing Systems. [Code]
    Jiannan Wu*, Muyan Zhong*, Sen Xing*, Zeqiang Lai*, Zhaoyang Liu*, Wenhai Wang*, Zhe Chen, Xizhou Zhu, Lewei Lu, Tong Lu, and  others

2023

  1. ControlLLM: Augment Language Models with Tools by Searching on Graphs
    ECCV 2024, European Conference on Computer Vision. [Code]
    Zhaoyang Liu*, Zeqiang Lai*, Gao Zhangwei, Erfei Cui, Zhiheng Li, Xizhou Zhu, Lewei Lu, Qifeng Chen, Yu Qiao, Jifeng Dai, and Wang Wenhai
  2. Mini-DALLE3: Interactive Text to Image by Prompting Large Language Models
    Preprint, arXiv preprint arXiv:2310.07653. [Website] [Code]
    Zeqiang Lai, Xizhou Zhu, Jifeng Dai, Yu Qiao, and Wenhai Wang
  4. InternGPT: Solving Vision-Centric Tasks by Interacting with ChatGPT Beyond Language
    Preprint, arXiv preprint arXiv:2305.05662. [Website] [Code]
    Zhaoyang Liu*, Yinan He*, Wenhai Wang*, Weiyun Wang*, Yi Wang*, Shoufa Chen*, Qinglong Zhang*, Zeqiang Lai*, Yang Yang, Qingyun Li, Jiashuo Yu, Kunchang Li, and 8 more authors
  5. Denoising Diffusion Semantic Segmentation with Mask Prior Modeling
    Preprint, arXiv preprint arXiv:2306.01721. [Code]
    Zeqiang Lai*, Yuchen Duan*, Jifeng Dai, Ziheng Li, Ying Fu, Hongsheng Li, Yu Qiao, and Wenhai Wang

2022

  1. Hybrid Spectral Denoising Transformer with Guided Attention
    ICCV 2023, International Conference on Computer Vision. [Poster] [Code]
    Zeqiang Lai, Chenggang Yan, and Ying Fu
  2. Mixed Attention Network for Hyperspectral Image Denoising
    Preprint, arXiv preprint arXiv:2301.11525. [Code]
    Zeqiang Lai, and Ying Fu
  3. Hyperspectral Image Super Resolution with Real Unaligned RGB Guidance
    TNNLS, IEEE Transactions on Neural Networks and Learning Systems. [Website] [Code]
    Zeqiang Lai, Ying Fu, and Jun Zhang

2021

  1. Deep plug-and-play prior for hyperspectral image restoration
    Neurocomputing, Elsevier Neurocomputing. [Code]
    Zeqiang Lai, Kaixuan Wei, and Ying Fu