Xingyao Wang (王星尧)

Pronouns: he/him/his. How to pronounce?

prof_pic.jpg

I am co-founder at All Hands AI All Hands AI, building open source software agent OpenHands Stargazers, and a CS PhD candidate at the UIUC University of Illinois Urbana-Champaign, advised by Prof. Heng Ji.

I am interested in AI that improves itself in the real world: in how a system can turn what people actually do with it into supervision that makes it better, and in building the full stack that makes this possible, from the product where those signals come from down to the model that learns from them.

I received my undergraduate degree at the University of Michigan in computer science and data science. Previously, I was an intern at Google Research (2023, multimodal LLM pre-training), Microsoft (2022, unit-test generation), ByteDance (2021, lightseq), and Tencent (2020, distributed deep learning).

selected publications

2026

  1. EvoClaw: Evaluating AI Agents on Continuous Software Evolution
    Gangda Deng, Zhaoling Chen, Zhongming Yu, Haoyang Fan, Yuhong Liu, Yuxin Yang, Dhruv Parikh, Rajgopal Kannan, Le Cong, Mengdi Wang, Qian Zhang, Viktor Prasanna, Xiangru Tang, and Xingyao Wang
    Arxiv preprint, Mar 2026
    Paper Website Code
  2. A Rubric-Supervised Critic from Sparse Real-World Outcomes
    Xingyao Wang, Valerie Chen, Heng Ji, and Graham Neubig
    Arxiv preprint, Mar 2026
    Paper Code Model Blog X/Twitter Thread
  3. The OpenHands Software Agent SDK: A Composable and Extensible Foundation for Production Agents
    Xingyao Wang, Simon Rosenberg, Juan Michelini, Calvin Smith, Hoang Tran, Engel Nyst, Rohit Malhotra, Xuhui Zhou, Valerie Chen, Robert Brennan, and Graham Neubig
    Proceedings of the Eighth Conference on Machine Learning and Systems, MLSys 2026, Bellevue, WA, USA (Oral), Nov 2026
    Paper

2025

  1. TOM-SWE: User Mental Modeling For Software Engineering Agents
    Xuhui Zhou, Valerie Chen, Zora Zhiruo Wang, Graham Neubig, Maarten Sap, and Xingyao Wang
    Arxiv preprint, Oct 2025
    Paper
  2. How can we assess human-agent interactions? Case studies in software agent design
    Valerie Chen, Rohit Malhotra, Xingyao Wang, Juan Michelini, Xuhui Zhou, Aditya Bharat Soni, Hoang H. Tran, Calvin Smith, Ameet Talwalkar, and Graham Neubig
    Arxiv preprint, Oct 2025
    Paper
  3. Devstral: Fine-tuning Language Models for Coding Agent Applications
    Abhinav Rastogi, Adam Yang, Albert Q. Jiang, Alexander H. Liu, Alexandre Sablayrolles, Amélie Héliou, Amélie Martin, Anmol Agarwal, Andy Ehrenberg, Andy Lo, Antoine Roux, Arthur Darcet, Arthur Mensch, Baptiste Bout, Baptiste Rozière, Baudouin De Monicault, Chris Bamford, Christian Wallenwein, Christophe Renaudin, Clémence Lanfranchi, Clément Denoix, Corentin Barreau, Darius Dabert Devon Mizelle, Diego Casas, Elliot Chane-Sane, Emilien Fugier, Emma Bou Hanna, Gabrielle Berrada, Gauthier Delerce, Gauthier Guinet, and 73 more authors
    Arxiv preprint, Aug 2025
    Paper
  4. LocAgent: Graph-Guided LLM Agents for Code Localization
    Zhaoling Chen, Xiangru Tang, Gangda Deng, Fang Wu, Jialong Wu, Zhiwei Jiang, Viktor Prasanna, Arman Cohan, and Xingyao Wang
    Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics, Mar 2025
    Paper Code

2024

  1. Training Software Engineering Agents and Verifiers with SWE-Gym
    Jiayi Pan*, Xingyao Wang*, Graham Neubig, Navdeep Jaitly, Heng Ji, Alane Suhr*, and Yizhe Zhang*
    ICML 2025; DL4C Workshop @ ICLR 2025, Dec 2024
    Paper Code X/Twitter Thread
  2. OpenHands: An Open Platform for AI Software Developers as Generalist Agents
    Xingyao Wang, Boxuan Li, Yufan Song, Frank F. Xu, Xiangru Tang, Mingchen Zhuge, Jiayi Pan, Yueqi Song, Bowen Li, Jaskirat Singh, Hoang H. Tran, Fuqiang Li, Ren Ma, Mingzhang Zheng, Bill Qian, Yanjun Shao, Niklas Muennighoff, Yizhe Zhang, Binyuan Hui, Junyang Lin, Robert Brennan, Hao Peng, Heng Ji, and Graham Neubig
    ICLR 2025, Jul 2024
    Paper Code
  3. Executable Code Actions Elicit Better LLM Agents
    Xingyao Wang, Yangyi Chen, Lifan Yuan, Yizhe Zhang, Yunzhu Li, Hao Peng, and Heng Ji
    ICML 2024; LLMAgents Workshop @ ICLR 2024 (Oral), Feb 2024
    Paper Code Slide Blog X/Twitter Thread Poster Demo
  4. A Single Transformer for Scalable Vision-Language Modeling
    Yangyi Chen*, Xingyao Wang*, Hao Peng, and Heng Ji
    Transactions on Machine Learning Research, 2024, Jul 2024
    Paper Code X/Twitter Thread
  5. MINT: Evaluating LLMs in Multi-turn Interaction with Tools and Language Feedback
    Xingyao Wang*, Zihan Wang*, Jiateng Liu, Yangyi Chen, Lifan Yuan, Hao Peng, and Heng Ji
    In Proceedings of the International Conference on Learning Representations, May 2024
    Paper Website Code Blog X/Twitter Thread Poster
  6. CRAFT: Customizing LLMs by Creating and Retrieving from Specialized Toolsets
    Lifan Yuan*, Yangyi Chen*, Xingyao Wang, Yi R. Fung, Hao Peng, and Heng Ji
    In Proceedings of the International Conference on Learning Representations, May 2024
    Paper Code
  7. LETI: Learning to Generate from Textual Interactions
    Xingyao Wang, Hao Peng, Reyhaneh Jabbarvand, and Heng Ji
    In Findings of the Association for Computational Linguistics: (NAACL); DaSH Workshop @ NAACL 2024 (Oral), Jun 2024
    Paper Code Slide X/Twitter Thread Poster

2023

  1. Code4Struct: Code Generation for Few-Shot Event Structure Prediction
    Xingyao Wang, Sha Li, and Heng Ji
    In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Jul 2023
    Paper Code X/Twitter Thread Poster