Hello! I am Zhongwei XIE, an MPhil student in HKUST. I am so fortune to be supervised by Prof. Yangqiu Song. I get the bachelor’s degree in Computer Science Elite Class from Wuhan university in June, 2025. I am interested in enhancing the long horizon cowork agentic and reasoning ability of LLMs and building efficient AI training framework. During my previous research journey, I am so lucky to work closely with Prof. Jiaxin Bai from HKBU, Prof. Wenhai Li from Wuhan University, Prof. Xiao Luo from UW–Madison and Dr. Jingru Yang from Carnegie Mellon University.
🔥 News
- 2026.06: I serve as session chair at ACL 2026.
- 2026.04: 1 manuscript is accept by ACL Findings 2026 (Topic: Data Selection For LLM Post Training/Self Evolving).
- 2025.11: I attended EMNLP 2025 in Nov 2025, Suzhou, China. Thanks for the support by Prof. Yangqiu Song.
- 2025.09: I start my Mphil journey in HKUST-KnowComp supervised by Prof. Yangqiu Song.
- 2025.06: I graduated from Wuhan University. I will miss everything.
- 2025.05: 1 manuscript is submitted to NIPS.
- 2025.03: I serve as a reviewer for ACL ARR.
- 2025.03: 🎉🎉 I start working together with Prof. Xiao Luo and Dr. Chong Chen as a research intern.
- 2025.01: 🎉🎉 I will join HKUST as a MPhil student (25 Fall), supervised by Prof. Yangqiu Song.
📝 Publications
CoworkEnv: Synthesizing Training Environments for Long-Horizon Cowork Agents
Zhongwei Xie, Yiming Du, Jiaxin Bai, Baixuan Xu, Chunyang Li, Xiyu REN, Wei Fan, Tianshi Zheng, Weizhi Du, Jingqiao Yu, Xingsu Li, Zhaowei Wang, Yangqiu Song
Preprint. HF daily paper #10 [paper]
JockeyBench: Evaluating LLM Agents on Horse-Racing Decision-Making
Zhongwei Xie*, Xingsu Li*, Jingqiao Yu*, Zihao Wang, Hong ChiTat, Zhaowei Wang, Weizhi Du, Yuxuan Liu, Huihao JING, Jiaxin Bai, Yangqiu Song
Preprint * Means equal contribution [paper]
A Survey on AI Agent Harness
Zhongwei Xie, Xiyu Ren, Tianshi Zheng, Jiaxin Bai, Wei Fan, Baixuan Xu, Haoran Li, Huihao Jing, Yangqiu Song
Preprint [paper]
NGDB-Zoo: Towards Efficient and Scalable Neural Graph Databases Training
Zhongwei Xie, Jiaxin Bai, Shujie LIU, Haoyu Huang, LI Yufei, Yisen Gao, Hong Ting Tsang, Yangqiu Song
NIPS 2026 [paper]
GALA: Geometric Data Selection with Strategic Prospecting for Large Language Model Self-training
Zhongwei Xie, Ruihao Liao, Zimo Wang, Chong Chen, Xian-Sheng Hua, Xiao Luo.
Findings of ACL 2026 [paper]
SpatialCrafter: Unleashing the Imagination of Video Diffusion Models for Scene Reconstruction from Limited Observations
Songchun Zhang, Huiyao Xu, Sitong Guo, Zhongwei Xie, Hujun Bao, Weiwei Xu, Changqing Zou
ICCV 2025. [paper]
🎖 Honors and Awards
- 2026.10 Invited Talk at Nice (Nexus for IntelligenCE) / Qingke AI / Huawei Hong Kong
- 2026.07 Research Travel Grant to ACL
- 2024.06 National First Prize, China Robot and Artificial Intelligence Competition
- 2024.05 Second prize, ASC Student Supercomputer Challenge
- 2023.11 Lei Jun Scholarship for Computer Science
- 2023.07 National First Prize, China Robot and Artificial Intelligence Competition
- 2023.06 1st place(Championship), RoboCup China Open @Home
📖 Educations
- 2025.09 - Now, Mphil, School of Computer Science, HKUST.
- 2021.09 - 2025.06, Undergraduate, School of Computer Science, Wuhan University.
💬 Miscellaneous
- I contributed 50% to the writing of a HK$1.1M research grant proposal, GRF (RGC), Scaling Low-Bit Training for Efficient Large Model Deployment.
- I maintain pictures on Instagram. Check them out.
- I enjoy Badminton🥍 Hiking 🚶♀️ Football ⚽️ in my free time.
- I am a fan of TV show 📺! I like Friends, Game of Thrones, Shameless and a lot more…
📖 Blogs/Books I read:
- I refer to many open source posts when gathering information and making decisions, thanks to the community.
- https://junweiliang.me/letter.html
- https://ysymyth.github.io/The-Second-Half/
- http://www.incompleteideas.net/IncIdeas/BitterLesson.html
- https://gwern.net/doc/psychiatry/depression/2024-hill.pdf
