OTULP

Building Perfect Self-Evolving Models Without Human Oversight
I am currently a Ph.D candidate in HCP Group at Sun Yat-sen University (SYSU), supervised by Prof. Xiaodan Liang. I received my Master's degree at the Sun Yat-sen University (SYSU), supervised by Prof. Xiaobin Chang. Prior to SYSU, I obtained a B.Eng from TYUT in 2021.

I have some research collaborations about the topic below. Feel free to contact me.
• Self-Improvement
• Continue/Lifelong/Increment Learning
• VLM / Gen Agent (skill; RL)
• Latent Reasoning

• [2026.07] 🔥 We release BEE, the detailed discrete version of EVA.
• [2026.06] 🎉 One paper (SyncLoop) (formerly C2-Evo) is accepted by ECCV 2026.
• [2026.02] 🎉 One paper (Atomicvla) is accepted by CVPR 2026.
• [2026.02] 🎉 One paper (SemHiTok) is accepted by ICLR 2026.
• [2025.12] 🔥 We release EVA, a framework that internalizes explicit tool calling into the model's intrinsic capabilities.
• [2025.07] 🔥 We release C2-Evo, a method for co-evolving models and data, an agent for drawing auxiliary line.
• [2025.02] 🔥 We release TransMamba, a simple method to quickly adapt Transformer models to the Mamba framework.
• [2023.10] 🎉 One paper is accepted by PRCV 2023.
• [2023.07] 🎉 One paper (DRC) is accepted by ICCV 2023.
• [2021.01] Welcome to my new homepage!

• arXiv 2026

  Arxiv Beyond the Eye: Efficient Multimodal Reasoning via Self-Regulated Implicit Visual Tools

  Xiuwei Chen, Quanlin Chen, Wentao Hu, Zisheng Chen, Kun Xiang, Zehua Ma, Mingyang Zhang, Jianhua Han, Hanhui Li, Hang Xu, Xiaodan Liang

  arXiv 2026

  PDF Code Project Page
• CVPR 2026

  CVPR 2026 AtomicVLA: Unlocking the Potential of Atomic Skill Learning in Robots

  Likui Zhang, Tao Tang, Zhihao Zhan, Xiuwei Chen, Zisheng Chen, Jianhua Han, Jiangtong Zhu, Pei Xu, Hang Xu, Hefeng Wu, Liang Lin, Xiaodan Liang

  CVPR 2026

  PDF Code Project Page
• ICLR 2026

  ICLR 2026 SemHiTok: A Unified Image Tokenizer via Semantic-Guided Hierarchical Codebook for Multimodal Understanding and Generation

  Zisheng Chen, Chunwei Wang, Runhui Huang, Hongbin Xu, Xiuwei Chen, Jun Zhou, Jianhua Han, Hang Xu, Xiaodan Liang

  ICLR 2026

  PDF
• arXiv 2025

  Arxiv Latent Visual States for Efficient Multimodal Reasoning

  Xiuwei Chen, Wentao Hu, Yongxin Wang, Zisheng Chen, Likui Zhang, Kun Xiang, Jianhua Han, Hui-Ling Zhen, Jingyuan Zou, Hang Xu, Xiaodan Liang

  arXiv 2025

  PDF
• ECCV 2026

  ECCV 2026 C2-Evo: Co-Evolving Multimodal Data and Model for Self-Improving Reasoning

  Xiuwei Chen, Wentao Hu, Hanhui Li, Jun Zhou, Zisheng Chen, Meng Cao, Yihan Zeng, Kui Zhang, Yu-Jie Yuan, Jianhua Han, Hang Xu, Xiaodan Liang

  ECCV 2026

  PDF Code Project Page
• arXiv 2025

  Arxiv TransMamba: Fast Universal Architecture Adaption from Transformers to Mamba

  Xiuwei Chen, Wentao Hu, Xiao Dong, Sihao Lin, Zisheng Chen, Meng Cao, Yina Zhuang, Jianhua Han, Hang Xu, Xiaodan Liang

  arXiv 2025

  PDF Code Project Page
• ICCV 2023

  ICCV 2023 Dynamic Residual Classifier for Class Incremental Learning

  Xiuwei Chen, Xiaobin Chang

  ICCV 2023

  PDF Code Project Page

• 2024.09 - now, PhD Student, Sun Yat-sen University (SYSU)
• 2021.09 - 2024.06, Master of CS, Sun Yat-sen University (SYSU)
• 2017.09 - 2021.06, Bachelor of CS, The Taiyuan University of Technology (TYUT)

• 2025.08 - now, Research Intern, YinWang 2030 Lab, Supervisor: Hang Xu, Jianhua Han
• 2024.07 - 2025.07, Research Intern, Huawei Noah's Ark Lab, Supervisor: Hang Xu, Jianhua Han, Yihan Zeng
• 2023.06 - 2023.08, Algorithm Intern, Hikvision Research Institute, Supervisor: Jun Che

© 2026 Xiuwei Chen. Last update: 2026.02.