Human-Centric Conversational AI with LLMs
By Yifei Yuan

Yifei Yuan will give a talk on Human-Centric Conversational AI with LLMs

Abstract

Large language models (LLMs) have transformed conversational AI, enabling systems to generate fluent, context-aware, and engaging dialogue. Human-centric conversational AI focuses on designing models that interact naturally with users, understand multimodal input, and adapt to diverse social contexts. This talk presents approaches for multimodal conversation modeling, social media understanding, and agentic AI, showing how LLMs can be leveraged to create interactive, contextually grounded, and engaging conversations. Examples illustrate the opportunities and challenges in building conversational systems that are both versatile and responsive, highlighting directions for the next generation of human-centered LLM-powered AI.

Biography

Yifei Yuan is a postdoctoral researcher at the Swiss Data Science Center, ETH Zurich, and a guest researcher (formerly a postdoc) at the Coastal NLP Group, University of Copenhagen. She received her PhD in 2023 from the Text Mining Group at The Chinese University of Hong Kong, under the supervision of Prof. Lam Wai. Her research lies at the intersection of natural language processing and information retrieval, with a focus on human-centered conversational AI, multimodal analysis of text and images, and social science applications. Her key areas include social media mining, multimodal conversational systems, agentic AI, and broader topics across the NLP and IR landscape.