Tianyu Bell Pan (潘天钰)
I am a Ph.D. candidate in the Department of Electrical and Computer Engineering (ECE) (Advisor: Dr. Damon Woodard) at the University of Florida, affiliated with Florida Institute for National Security (FINS) and Applied Artificial Intelligence Group (AAIG). I hold an MS degree in ECE and MA degree in Economics (IO) from University of Florida.
My research focuses on multimodal learning, particularly in computer vision. This includes areas such as training (primarily pre-training and structured representation learning), information retrieval, optimization, and reinforcement learning. The tools I utilize are mainly from information theory, control theory, and geometric measure theory.
Office: 206 D1, Materials Engineering Building (UF FINS), Herbert Wertheim College of Engineering, University of Florida, Gainesville, FL, 32611
Email: tpan1@ufl.edu
News
12/2025: From Kakeya to Kernels: A Multi-Scale Geometric Framework for Robust Representation Learning is available on ArXiv.
10/2025: Psychology of Phishing Emails: Quantifying Persuasion Principles and Simulating Detection with Large Language Models is released on SSRN.
09/2025: Lyapunov-Stable Adaptive Control for Multimodal Concept Drift is accepted by the 39th Conference on Neural Information Processing Systems (NeurIPS 2025).
08/2025: From Syntax to Semantics: Evaluating the Impact of Linguistic Structures on LLM-Based Information Extraction is available in Proceedings of the 1st Joint Workshop on Large Language Models and Structure Modeling (XLLM 2025).
07/2025: Efficient Generalization via Multimodal Co-Training under Data Scarcity and Distribution Shift is presented in the 42nd ICML Conference Affinity Workshop New In ML.
01/2025: LLM4RE: A Data-centric Feasibility Study for Relation Extraction is presented in COLING 2025.
Tianyu Bell Pan 