Weili Cao

I am a Computer Science PhD student at Duke, co-advised by Prof. Shuyan Zhou and Prof. Bhuwan Dhingra. I completed my Master's and Undergraduate degrees at UCSD, where I was fortunate to be advised by Dr. Leon Bergen and Dr. Ramamohan Paturi.

My research focuses on improving Large Language Models’ (LLMs) task-solving capabilities through the lens of information retrieval (IR), extraction (IE), and utilization. I study how LLMs can better process and utilize information at inference time, especially when relevant evidence is scattered across long contexts and exhibits complex interdependencies.

I am currently working on long-context modeling for LLM agents.