Title: Revealing and Mitigating Social Biases of LLMs via Moral Reasoning

Speaker: Prof. Ruihong Huang @ Texas A&M University

Time : 9:00 - 10:00, June 4th, 2026

Location: Online: https://hli.skku.edu/InvitedTalk260604 

Language: English speech & English slides

Abstract: 

Despite the remarkable capabilities in a multitude of NLP tasks, pre-trained large language models (LLMs) frequently manifest various social biases that reflect or amplify stereotypes found in their training data. In this talk, I will use gender bias as an example and present our recent work that engages in moral reasoning for revealing implicit biases of LLMs as well as bias mitigation by fostering moral exploratory thinking. If time permits, I would also like to present our recent research progress on evaluating LLM's dialogue understanding capabilities and accelerating GRPO-style reinforcement learning through diversity aware reward reweighting.

Bio:

Dr. Ruihong Huang is an associate professor in the Department of Computer Science & Engineering at Texas A&M University (TAMU), College Station. She is also an adjunct associate professor in McWilliams School of Biomedical Informatics at UTHealth Houston. Huang received her PhD in computer science at the University of Utah and completed a postdoc at Stanford University. Her research is focused on event-centric NLP,  discourse analysis, dialogue and pragmatics, LLM evaluation, safety and moral reasoning of LLMs. She is a recipient of the US National Science Foundation CAREER award (2020).