About Me

I am a PhD student at the National University of Singapore, fortunate to be supervised by Prof. HUANG Zhiyong. My research is primarily focused on multimodality, autonomous agents, and code generation.

Previously, I was a research intern at Qwen, supervised by Junyang Lin and Binyuan Hui, where I contributed to building the Qwen model series, including Qwen-VL, Qwen-3.5, and Qwen-Coder. I have also had the great opportunity to work with Prof. Tat-Seng Chua, Prof. Junxian He, and Prof. Michael Qizhe Shieh.

Recent News

  • Apr 2026: One paper accepted to ICML 2026.
  • Feb 2026: Released Qwen-3.5 and Qwen 3 Coder Next.
  • Jan 2026: Three papers accepted to ICLR 2026.
  • Sep 2025: Released Qwen3-VL.
  • Sep 2025: One paper (SE-GUI) accepted to NeurIPS 2025.
  • Aug 2025: Two papers accepted to EMNLP 2025.
  • Jul 2025: ScreenSpot-Pro accepted to ACMMM 2025 as an Oral Presentation.
  • May 2025: Three papers accepted to ACL 2025.

Selected Publications

Agents & GUI Interaction

ScreenSpot-Pro: GUI Grounding for Professional High-Resolution Computer Use (Oral Presentation)
Kaixin Li, Ziyang Meng, Hongzhan Lin, Ziyang Luo, Yuchen Tian, Jing Ma, Zhiyong Huang, Tat-Seng Chua
ACMMM 2025 · ICLR 2025 Workshop on Reasoning and Planning for LLMs
[Paper] [Code] [Dataset]
Used by Google Gemini, GPT-5.2, Qwen-VL Series, Microsoft OmniParser, and Meta's Muse Spark.
65,000+ dataset downloads.
Enhancing Visual Grounding for GUI Agents via Self-Evolutionary Reinforcement Learning
Xinbin Yuan, Jian Zhang, Kaixin Li, Zhuoxuan Cai, Jie Chen, Lujian Yao, Enguang Wang, Qibin Hou, Jinwei Chen, Peng-Tao Jiang, Bo Li
NeurIPS 2025
[Paper] [Code] [Models]
Grounding Computer Use Agents on Human Demonstrations
Aarash Feizi, Shravan Nayak, Xiangru Jian, Kevin Qinghong Lin, Kaixin Li, Rabiul Awal, Xing Han Lù, Johan Obando-Ceron, Juan A. Rodriguez, Nicolas Chapados, David Vazquez, Adriana Romero-Soriano, Reihaneh Rabbany, Perouz Taslakian, Christopher Pal, Spandana Gella, Sai Rajeswara
ICLR 2026
[Paper]
HackWorld: Evaluating Computer-Use Agents on Exploiting Web Application Vulnerabilities
Xiaoxue Ren*, Penghao Jiang*, Kaixin Li*, Zhiyong Huang, Xiaoning Du, Jiaojiao Jiang, Zhenchang Xing, Jiamou Sun, Terry Yue Zhuo
ICLR 2026
[Paper]
Robi Butler: Remote Multimodal Interactions with Household Robot Assistant
Anxing Xiao, Nuwan Janaka, Tianrun Hu, Anshul Gupta, Kaixin Li, Cunjun Yu, David Hsu
ICRA 2025
[Paper] [Project Page]

Coding & Code Intelligence

MMCode: Evaluating Multi-Modal Code Large Language Models with Visually Rich Programming Problems
Kaixin Li, Yuchen Tian, Qisheng Hu, Ziyang Luo, Zhiyong Huang, Jing Ma
EMNLP 2024 Findings · ICLR 2025 Workshop on Reasoning and Planning for LLMs
[Paper] [Code] [Dataset]
Used by Qwen2.5-VL and Qwen3-VL.
BigCodeArena: Unveiling More Reliable Human Preferences in Code Generation via Execution
Terry Yue Zhuo, Xiaolong Jin, Hange Liu... Kaixin Li, Yuhan Cao, Bo Liu, et al.
arXiv
[Paper] [Code]
Tree-of-Evolution: Tree-Structured Instruction Evolution for Code Generation in Large Language Models
Ziyang Luo, Kaixin Li, Hongzhan Lin, Yuchen Tian, Mohan Kankanhalli, Jing Ma
ACL 2025
[Paper]
MCTS-Judge: Test-Time Scaling in LLM-as-a-Judge for Code Correctness Evaluation
Yutong Wang, Pengliang Ji, Chaoqun Yang, Kaixin Li, Ming Hu, Jiaoyang Li, Guillaume Sartoretti
arXiv preprint (2502.12468)
[Paper]
InstructCoder: Empowering Language Models for Code Editing (Oral Presentation)
Kaixin Li*, Qisheng Hu*, Xu Zhao, Yuxi Xie, Tiedong Liu, Hui Chen, Qizhe Xie*, Junxian He*
ACL 2024 SRW
[Paper] [Code] [Dataset]

Open Source

  • TACO-verified: Verified code contest problems and solutions.
    5,000+ downloads · Widely used by the community and by DeepCoder.
    [Dataset]
  • IconStack-48M: The largest open-source icon dataset with 48 million images, SVGs, captions, and rich metadata.
    [Dataset]

Services

Peer Reviewer for:

  • ICLR (2024, 2025, 2026)
  • CVPR (2025)
  • ECCV (2026)
  • ACL (2025)
  • EMNLP (2025)
  • AAAI (2025)
  • ACMMM (2025)