Lifu Huang

Computer Science Department,

UC Davis

Email: lfuhuang@ucdavis.edu

Davis, CA, US

Google Scholar

Github

Twitter



My research centers on Natural Language Processing, Machine Learning and Artificial Intelligence. Our current research focuses on vision-language models, agentic AI, and the robustness of RL-based post-tuning. We are particularly interested in building multimodal AI systems that can better understand and interact with the physical world, while also studying issues such as reward hacking and other alignment failures to make post-trained models more reliable and trustworthy. We are grateful to NSF, DARPA, IARPA, U.S. Air Force, Amazon (AWS and Alexa AI), Meta AI, Google, Intuit and Washington Post for supporting our research!

Ph.D. Students

  • Menglong (Barry) Yao: Continual Adaptation of Multimodal Foundation Models, (Fall 2022)
  • Zihao Lin: Multimodal and Agentic AI, (Fall 2023)
  • Mohammad Beigi: Reward Hacking of RL-based Post-Tuning; Uncertainty Estimation and Calibration for LLMs, (Fall 2023)
  • Haibo Wang: 3D Understanding and Spacial Intelligence with Multimodal Foundation Models, (Fall 2025)

MS Students

  • Muyang Zheng (MS from UC Davis): Agentic AI, Multimodal Reasoning

Visiting Students

  • Ying Shen (PhD from UIUC): Multimodal Learning, Embody AI, (2021 - 2025)
  • Yuexi Shen (MS from UCSB): Novelty Evaluation for Scientific Hypotheses; Strategic Conversation Simulation with RL; (2025-2026)

Alumi

  • Zhiyang Xu (PhD, 2021-2026), now a Research Scientist at Institute of Foundation Models. Thesis topic: Towards Unified and Generalizable Multimodal Foundation Models
  • Minqian Liu (PhD, 2021-2026), now a Research Scientist at Microsoft AI. Thesis topic: Holistic and Generalizable Evaluation of Generative Models
  • Jingyuan Qi (PhD, 2022-2026), now a Research Scientist at Eigen AI. Thesis topic: Knowledge-Centric Multimodal Intelligence: Understanding, Reasoning, and Generation
  • Sijia Wang (PhD, 2020-2024), now a Research Scientist at Amazon AWS AI. Thesis topic: Towards Generalizable Information Extraction with Limited Supervision
  • Zoe Zheng (MS from VT, co-advised with Chris Thomas, 2022-2024). Thesis topic: Advancing Chart Question Answering with Robust Chart Component Recognition
  • Tong Zhou (MS from VT, 2022-2024)
  • Xiaochu Li (MS from VT, 2021-2023). Thesis topic: GlitchAgent: Detecting Video Game Glitches from Gameplay Videos
  • Trevor Ashby (MS from VT, 2023-2024). Thesis topic: Towards Effective Long Conversation Generation: Dynamic Topic Tracking and Recommendation for Open-Domain Dialogue Systems
  • Sai Gurrapu (MS from VT, co-advised with Feras Batarseh, 2022). Thesis topic: Explainable Neural Claim Verification Using Rationalization
  • Pei Wang (MS from VT, co-advised with Jin-Hee Cho, 2021-2022). Thesis topic: Generative Chatbot Framework for Cybergrooming Prevention