Bio

I am a Postdoctoral Scholar with Professor James Zou at Stanford University. I received my Ph.D. in Computer Science from UCLA, where I was a member of the UCLA Natural Language Processing Group (UCLA NLP) and the Center for Vision, Cognition, Learning, and Autonomy (VCLA). Previously, I earned my M.S. in Computer Science at Tsinghua University. My research has been funded by the Amazon Ph.D. Fellowship, Bloomberg Data Science Ph.D. Fellowship, Qualcomm Innovation Fellowship, UCLA Dissertation Year Fellowship, and the NeurIPS Scholar Award.

Honors & Awards

  • Bloomberg Data Science Ph.D. Fellowship, Bloomberg (2023)
  • Qualcomm Innovation Fellowship, Qualcomm (2023)
  • UCLA Dissertation Year Fellowship, UCLA (2023)
  • Amazon PhD Fellowship, Amazon (2023)
  • Outstanding Master Thesis Award, Tsinghua University (2018)

Boards, Advisory Committees, Professional Organizations

  • Program Chair, The 4th Southern California Natural Language Symposium (SoCal NLP) (2023 - 2023)
  • Area Chair, The Thirteenth International Conference on Learning Representations (ICLR 2025) (2024 - Present)
  • Co-organizer, The 4th Workshop on MATH-AI at NeurIPS 2024 (2024 - 2024)
  • Co-organizer, Workshop on AI for Math at ICML 2024 (2024 - 2024)
  • Co-organizer, Workshop on Tool-Augmented VIsion (TAVI) at CVPR 2024 (2024 - 2024)
  • Co-organizer, The 3rd Workshop on MATH-AI at NeurIPS 2023 (2023 - 2023)
  • Co-organizer, The 2nd Workshop on MATH-AI at NeurIPS 2022 (2022 - 2022)
  • Co-organizer, Workshop on Math AI for Education at NeurIPS 2021 (2021 - 2021)
  • Co-organizer, Tutorial on Deep Learning in Mathematical Reasoning at IJCAI 2023 (2023 - 2023)

Professional Education

  • Ph.D., UCLA, Computer Science (2024)

Stanford Advisors

Contact

  • Academic
    University - Scholar Department: Department of Biomedical Data Science Position: Postdoctoral Scholar

Additional Info

  • Mail Code: 5464

Links

Current Research and Scholarly Interests

My research goal is to build machines that can reason and collaborate with humans for the common good. My primary research focuses on machine learning and NLP, particularly machine reasoning, mathematical reasoning, and scientific discovery:
1. Mathematical reasoning in multimodal and knowledge-intensive contexts
2. Tool-augmented large language models for planning, reasoning, and generation
3. Parameter-efficient fine-tuning for fondation models
4. AI for scientific reasoning and discovery

All Publications

  • Optimizing generative AI by backpropagating language model feedback. Nature Yuksekgonul, M., Bianchi, F., Boen, J., Liu, S., Lu, P., Huang, Z., Guestrin, C., Zou, J. 2025; 639 (8055): 609-616

    Abstract

    Recent breakthroughs in artificial intelligence (AI) are increasingly driven by systems orchestrating multiple large language models (LLMs) and other specialized tools, such as search engines and simulators. So far, these systems are primarily handcrafted by domain experts and tweaked through heuristics rather than being automatically optimized, presenting a substantial challenge to accelerating progress. The development of artificial neural networks faced a similar challenge until backpropagation and automatic differentiation transformed the field by making optimization turnkey. Analogously, here we introduce TextGrad, a versatile framework that performs optimization by backpropagating LLM-generated feedback to improve AI systems. By leveraging natural language feedback to critique and suggest improvements to any part of a system-from prompts to outputs such as molecules or treatment plans-TextGrad enables the automatic optimization of generative AI systems across diverse tasks. We demonstrate TextGrad's generality and effectiveness through studies in solving PhD-level science problems, optimizing plans for radiotherapy treatments, designing molecules with specific properties, coding, and optimizing agentic systems. TextGrad empowers scientists and engineers to easily develop impactful generative AI systems.

    View details for DOI 10.1038/s41586-025-08661-4

    View details for PubMedID 40108317

    View details for PubMedCentralID 10794143

  • VISCO: Benchmarking Fine-Grained Critique and Correction Towards Self-Improvement in Visual Reasoning Wu, X., Ding, Y., Li, B., Lu, P., Yin, D., Chang, K., Peng, N., IEEE COMPUTER SOC IEEE COMPUTER SOC. 2025: 9527-9537

    View details for DOI 10.1109/CVPR52734.2025.00890

    View details for Web of Science ID 001601106700307