Lorraine Li

Lorraine Li is an Assistant Professor in the Department of Computer Science at the University of Pittsburgh, affiliated with the Interdisciplinary Science Program (ISP) and the School of Computing and Information. She holds a PhD from the University of Massachusetts Amherst (2022) and conducted postdoctoral research at AI2's Mosaic team. Her work focuses on NLP, machine learning, and socially responsible AI systems.

• Education: PhD in Computer Science (UMass Amherst, 2022)

Research explores evaluation frameworks for commonsense knowledge, model interpretability, and ethical AI applications in domains like education and law. Key interests include probabilistic models, long-tail reasoning, and geographic robustness in LLMs.

Recent publications address confirmation bias in reasoning chains (ACL 2025), geographically diverse prompting (CVPR 2024), and uncommon scenario reasoning (NAACL 2024). She co-organized the AAAI 2024 Make symposium and serves on committees for ACL, EMNLP, and NAACL.

• Grants: Pitt Cyber funding (2024)
• Lab: Pitt NLP Seminar group