Guowei Xu

I am an undergraduate student at Yao Class, Tsinghua University. I am interested in language models and reinforcement learning.

Currently, I am a research intern at Stanford AI for Science Lab (Prof. James Zou) and Harvard Embodied Minds Lab (Prof. Yilun Du). Previously, I was fortunate to intern at Tsinghua Embodied AI Lab (Prof. Huazhe Xu) and PKU-YuanGroup (Prof. Li Yuan).

I am currently actively seeking PhD opportunities for Fall 2027. I am always eager to connect with fellow researchers and explore potential collaborations or exchange ideas.

Education

IIIS, Tsinghua Univeristy
Undergraduate Student
2023 - now
GPA: 4.0/4.0, with 17 A+ (Top ~5%) and 15 A (Top ~25%)

Publications & Preprints

Papers sorted by recency. Representative papers are highlighted.

Sparse Reward Subsystem in Large Language Models
Guowei Xu, Mert Yuksekgonul, James Zou
arXiv , 2026
arXiv / bibtex

metaTextGrad: Automatically optimizing language model optimizers
Guowei Xu, Mert Yuksekgonul, Carlos Guestrin, James Zou
Conference on Neural Information Processing Systems (NeurIPS), 2025
arXiv / code / bibtex

LLaVA-CoT: Let Vision Language Models Reason Step-by-Step
Guowei Xu*, Peng Jin*, Ziang Wu*, Hao Li, Yibing Song, Lichao Sun, Li Yuan
International Conference on Computer Vision (ICCV), 2025
model & dataset / arXiv / code / bibtex

MENTOR: Mixture-of-Experts Network with Task-Oriented Perturbation for Visual Reinforcement Learning
Suning Huang*, Zheyu Zhang*, Tianhai Liang, Yihan Xu, Zhehao Kou, Chenhao Lu, Guowei Xu, Zhengrong Xue, Huazhe Xu
International Conference on Machine Learning (ICML), 2025
project page / arXiv / code / bibtex

ACE: Off-Policy Actor-Critic with Causality-Aware Entropy Regularization
Tianying Ji*, Yongyuan Liang*, Yan Zeng, Yu Luo, Guowei Xu, Jiawei Guo, Ruijie Zheng, Furong Huang, Fuchun Sun, Huazhe Xu
International Conference on Machine Learning (ICML), 2024 (Oral)
project page / arXiv / code / bibtex

DrM: Mastering Visual Reinforcement Learning through Dormant Ratio Minimization
Guowei Xu*, Ruijie Zheng*, Yongyuan Liang*, Xiyao Wang, Zhecheng Yuan, Tianying Ji, Yu Luo, Xiaoyu Liu, Jiaxin Yuan, Pu Hua, Shuzhen Li, Yanjie Ze, Hal Daumé III, Furong Huang, Huazhe Xu
International Conference on Learning Representations (ICLR), 2024 (Spotlight)
project page / arXiv / code / bibtex

Can Pre-Trained Text-to-Image Models Generate Visual Goals for Reinforcement Learning?
Jialu Gao*, Kaizhe Hu*, Guowei Xu, Huazhe Xu
Conference on Neural Information Processing Systems (NeurIPS), 2023
project page / arXiv / code / bibtex

Selected open-source projects

LLaVA-CoT Official Implementation⭐ 2123
Guowei Xu, Peng Jin, Ziang Wu (Code Implementation)
Nov 2024
model & dataset / arXiv / code / bibtex

DrM Official Implementation⭐ 78
Guowei Xu, Ruijie Zheng, Yongyuan Liang (Code Implementation)
Sept 2023
project page / arXiv / code / bibtex

Selected Talk and Reports

YouTuber Reports: LLaVA-CoT: Let Vision Language Models Reason Step-by-Step
Speaker
Timestamp 7:58 (YouTube), 2024
video link

Invited Talk: Mastering Visual Reinforcement Learning through Dormant Ratio Minimization
Speaker
88th Seminar (RL China), 2024
video link

Invited Talk: Mastering Visual Reinforcement Learning through Dormant Ratio Minimization
Speaker
ICLR Seminar (AI TIME), 2024
video link

Selected Awards

Sparking Program
The most prestigious and selective academic organization for students at THU
Tsinghua University (THU), 2025 (Top 1% in Tsinghua)

National Scholarship
scholar
China, 2024, 2025 (Top Scholarship in China)

Freshman Scholarship (First Prize)
scholar
Tsinghua University (THU), 2024 (Top Scholarship for Freshman)

The 52nd International Physics Olympiad (IPhO 2022)
Overall Winner (best total score)
International Physics Olympiad (IPhO), 2022

Teaching Assistant

Deep Reinforcement Learning (Graduate Course)
Director: Huazhe Xu
TAs: Guowei Xu, Kaizhe Hu, Pu Hua
Tsinghua University (THU), 2024

Natural Language Processing
Director: Tianxing He
TAs: Guowei Xu, Yeqi Feng, Minrui Luo
Tsinghua University (THU), 2025

Reviewer

International Conference on Learning Representations (ICLR)
Role: Reviewer for main conference
Official Site: https://iclr.cc/
Years: 2025, 2026

International Conference on Machine Learning (ICML)
Role: Reviewer for main conference
Official Site: https://icml.cc/
Years: 2025, 2026

Conference on Neural Information Processing Systems (NeurIPS)
Role: Reviewer for main conference
Official Site: https://nips.cc/
Years: 2025

Conference on Neural Information Processing Systems (NeurIPS)
Role: Volunteer
Official Site: https://nips.cc/
Years: 2025

Conference on Robot Learning (CoRL)
Role: Reviewer for workshop
Official Site: https://www.corl.org/
Years: 2023

IEEE Transactions on Multimedia
Role: Reviewer
Official Site: https://ieeexplore.ieee.org/xpl/RecentIssue.jsp?punumber=6046
Years: 2025

Contact

You are very welcome to contact me. I typically respond within a few days.
I can be contacted directly at xgw23 [at] mails.tsinghua.edu.cn