معرفی
Zhichao Xu is a researcher at AWS Bedrock AI, specializing in understanding language models' mechanisms and applications. He holds a PhD in Computer Science from the University of Utah, advised by Prof. Vivek Srikumar and Prof. Bei Wang. His work focuses on deep learning methods in NLP and Information Retrieval, including retrieval-augmented generation, in-context learning, and decoding techniques.
Education:
- PhD in Computer Science, University of Utah (2020-2024)
- MS in Computer Science, Rutgers University (2018-2020)
- BS in Applied Statistics, Shanghai University of Finance and Economics (2014-2018)
Research Interests: Zhichao explores the intersection of NLP and IR, emphasizing robust and interpretable systems. Key areas include: retrieval-augmented generation, LLM compression, ethical AI, and explainable recommendation. His work bridges theoretical advancements with practical applications in search, dialogue, and fairness.
Papers Trends: Recent publications highlight innovations in model architectures (e.g., CSPLADE for sparse retrieval), safety evaluation of compressed LLMs, and novel applications of state space models for text reranking. He also contributes to foundational surveys like the A Survey of Model Architectures in Information Retrieval.
Awards:
- Distinguished Paper Award at ISSTA 2023
- Honorable Mention at SIGIR-AP 2023
Teaching & Service: Taught Machine Learning, Data Mining, and Discrete Structures at the University of Utah. Served as a reviewer for top conferences (ACL, EMNLP) and journals (TOIS). Active in organizing workshops on LLM interpretability and fairness.
