Training-Free Bayesianization for Low-Rank Adapters of Large Language Models
Advances in Neural Information Processing Systems (NeurIPS), 2025
haizhou-shi
Professional Summary
I am an incoming Ph.D. student in the Computer Science Department at Rutgers University. I received my Bachelor’s degree at Huazhong University of Science and Technology in 2024. I was under the guidance of Prof. Kun He @ HUST, Prof. Hao Wang @ Rutgers and Prof. Huan Zhang @ UIUC.
From such a gentle thing, from such a fountain of all delight, my every pain is born.
—— Michelangelo
Education
Ph.D. Computer Science
2026-09-01
Present
Rutgers University
B.E. Computer Science
2019-09-01
2024-05-31
Huazhong University of Science and Technology
Interests
LLM thinking
Bayesian LLMs
Trustworthy AI
📚 My Research
I aim to enhance the reasoning capabilities of LLMs through efficient and streamlined methodologies. I also leverage Bayesian approaches to improve LLM trustworthiness and reliability.
Please feel free to reach out—I’m always open to collaboration! 😃
Featured Publications
•
•
1 min read
BLoB: Bayesian Low-Rank Adaptation by Backpropagation for Large Language Models
Advances in Neural Information Processing Systems (NeurIPS), 2024
Robustness-Aware Word Embedding Improves Certified Robustness to Adversarial Word Substitutions
Findings of ACL 2023
Recent Publications
(2025).
Improving Data Efficiency for LLM Reinforcement Fine-tuning Through Difficulty-targeted Online Data Selection and Rollout Replay.
NeurIPS 2025.
(2025).
TokUR: Token-Level Uncertainty Estimation for Large Language Model Reasoning.
ICLR 2026.
(2024).
Training-Free Bayesianization for Low-Rank Adapters of Large Language Models.
NeurIPS 2025.
(2024).
BLoB: Bayesian Low-Rank Adaptation by Backpropagation for Large Language Models.
NeurIPS 2024.
(2024).
Continual learning of large language models: A comprehensive survey.
ACM Computing Surveys.
Blogs