PhD student · Human–Computer Interaction · University of Houston

Hi, I’m Pei-Chi.

I’m a second-year PhD student at the University of Houston, advised by Emani Hicks in the AWARE Lab. I’m interested in human-centered AI for health and social wellbeing, both developing the models and designing how people interact with them through wearables and apps. My research interests include:

Human-centered AI for health and social wellbeing
Human–AI interaction with LLMs and ML systems
Community-centered design of intelligent technology

See my research Download CV

Current project

Study runs fall 2026 · 2 weeks per participant

EarMark👂

A Multimodal Dataset of Everyday Noise-Sensitivity Responses from Self‑Report and Apple Watch Sensing

Our long-term goal is an AI system that can anticipate difficult moments for people with noise sensitivity, such as misophonia and hyperacusis, and help them cope in everyday life. Right now we're building the foundation: participants wear an Apple Watch and answer short phone check-ins, so we can learn which everyday contexts tend to trigger a reaction. This data will become the basis for training personalized prediction models. No audio or location is ever stored.

Publications

TMLR 2026
Reward Modeling for Reinforcement Learning-Based LLM Reasoning: Design, Challenges, and Evaluation Pei-Chi Pan, Yingbin Liang, Sen Lin
NeurIPS 2026
Specialists Hold, Generalists Discount: Asymmetric Equilibrium in LLM Routing Auctions Hou, Lu, Karanjai, Pan, Lin, Xu, Shi
Under review
Anchors, Mixed Equilibria, and Welfare in LLM Routing Auctions Hou, Lu, Karanjai, Pan, Lin, Xu, Shi