Zhenxuan Fan

Zhenxuan Fan

Master’s student

Hi, I'm Zhenxuan Fan (θŒƒζŒ―η’‡) 🍻

δ½ ε₯½! / Hello! / μ•ˆλ…•ν•˜μ„Έμš”! / こんにけは! / Здравствуй! / Bonjour! / Guten Tag! / Hola! / Ciao! / Ψ§Ω„Ψ³Ω„Ψ§Ω… ΨΉΩ„ΩŠΩƒΩ…!

I am a first-year master’s student in the School of Software Technology, Zhejiang University, under the supervision of Academician Beng Chin Ooi and Researcher Wenqiao Zhang. Prior to this, I received my bachelor’s degree from the Software Colleage, Northeastern University (China).

I am currently conducting research in Embodied Intelligence, especially on VLA models. In the past, I focused on the efficient training and inference of LLMs and MLLMs, including model quantization, parameter-efficient fine-tuning and CoT compression.

πŸ‘‹πŸ‘‹πŸ‘‹ If you’re interested in my work, please feel free to reach out for discussions or collaborations!

Contact me via:
πŸ“§ Mail: zxfan[AT]zju.edu.cn / fzx0228[AT]gmail.com

🌈 What's new:

  • [06/2026] πŸš— Internship: I joined XPENG Robotics as a Research Intern, working on Pre-Training of VLA Foundation Models! πŸ€–
  • [05/2026] πŸŽ‰ CVPR'26: We won the championship at the CVPR 2026 ManipArena Challenge, and AgentVLA has been accepted to the CVPR 2026 EAI Workshop! πŸ”₯πŸ”₯πŸ”₯
  • [04/2026] πŸŽ‰ ACL'26: CoMoL has been accepted to ACL 2026 Findings! πŸŽ‰
  • [02/2026] πŸ“œ Preprint: We released Draft-Thinking.
  • [01/2026] πŸ“œ Preprint: We released CtrlCoT.
  • [01/2026] πŸŽ‰ ICLR'26: OmniCT has been accepted to ICLR 2026! πŸŽ‰
  • [01/2026] πŸŽ‰ ICASSP'26: FAQ has been accepted to ICASSP 2026! πŸŽ‰
  • [11/2025] πŸŽ‰ AAAI'26: MAU-GPT has been accepted to AAAI 2026! πŸŽ‰
  • [09/2025] πŸŽ“ Starting Master’s@ZJU: I will join Academician Beng Chin Ooi's group at SST, ZJU in 2025 Fall!
  • [06/2025] πŸŽ“ Graduation@NEU: Received my Bachelor's degree from Northeastern University as an Outstanding Graduate.

πŸŽ“ Selected Publications (Google Scholar)

CoMoL: Efficient Mixture of LoRA Experts via Dynamic Core Space Merging.
Jie Cao*, Zhenxuan Fan*, Zhuonan Wang, Tianwei Lin, Ziyuan Zhao, Rolan Yan, Wenqiao Zhang, Feifei Shao, Hongwei Wang, Jun Xiao, Siliang Tang
arXiv preprint, 2026.
[PDF] [arXiv] [bibtex]
Draft-Thinking: Learning Efficient Reasoning in Long Chain-of-Thought LLMs.
Jie Cao*, Tianwei Lin*, Zhenxuan Fan, Bo Yuan, Ziyuan Zhao, Rolan Yan, Wenqiao Zhang, Siliang Tang
arXiv preprint, 2026.
[PDF] [arXiv] [bibtex]
CtrlCoT: Dual-Granularity Chain-of-Thought Compression for Controllable Reasoning.
Zhenxuan Fan, Jie Cao, Yang Dai, Zheqi Lv, Wenqiao Zhang, Zhongle Xie, Peng Lu, Beng Chin Ooi
arXiv preprint, 2026.
[PDF] [arXiv] [bibtex]
OmniCT: Towards a Unified Slice-Volume LVLM for Comprehensive CT Analysis.
Tianwei Lin*, Zhongwei Qiu*, Wenqiao Zhang, Jiang Liu, Yihan Xie, Mingjian Gao, Zhenxuan Fan, Zhaocheng Li, Sijing Li, Zhongle Xie, Peng Lu, Yueting Zhuang, Ling Zhang, Beng Chin Ooi, Yingda Xia
The Fourteenth International Conference on Learning Representations, 2026.
[OpenReview] [PDF] [arXiv] [bibtex]
Enhancing Post-Training Quantization via Future Activation Awareness.
Zheqi Lv*, Zhenxuan Fan*, Qi Tian, Wenqiao Zhang, Yueting Zhuang
arXiv preprint, 2026.
[PDF] [arXiv] [bibtex]
MAU-GPT: Enhancing Multi-type Industrial Anomaly Understanding via Anomaly-aware and Generalist Experts Adaptation.
Zhuonan Wang*, Zhenxuan Fan*, Siwen Tan*, Yu Zhong, Yuqian Yuan, Haoyuan Li, Hao Jiang, Wenqiao Zhang, Feifei Shao, Hongwei Wang, others
Proceedings of the AAAI Conference on Artificial Intelligence, 2026.
[PDF] [arXiv] [bibtex]

πŸ§‘β€πŸ’» Research Intern

XPENG Robotics

Embodied Intelligence Algorithm Intern

XPENG Robotics Β· Shenzhen Β· June 2026 – Present

Topic: Pre-training and mid-training of VLA foundation models.

πŸ† Honors and Awards

Site Analytics