Towards Building Next-Generation AI Assistants
Date: 2024/07/30 - 2024/07/30
Academic Seminar: Towards Building Next-Generation AI Assistants
Speaker: Fanghua Ye, Ph.D.Candidate, University College London
Time: 9:00-10:00 a.m., July 30, 2024 (Beijing Time)
Location: Room 454, JI Long Bin Building
Abstract
As technology continues to advance, the incorporation of artificial intelligence (AI) into our daily lives has become increasingly prevalent. One notable manifestation of this trend is the emergence of AI assistants, which serve a myriad of roles - from virtual personal assistants managing routine tasks, to social companion chatbots for entertainment, to digital consultants for recommendations. Yet, ensuring the effectiveness, practicability, and safety of these AI assistants remains a critical challenge. They may produce irrelevant, non-factual, hallucinated, impolite, and even harmful responses, posing significant risks to their deployment in real-world applications. In this talk, I will share my past research on building effective AI assistants, with a particular focus on accurately modeling user intents and states. The underlying principle is that only by accurately capturing user needs can these AI assistants take appropriate actions and deliver suitable responses. I will also discuss my ongoing research regarding how large language models (LLMs) can be leveraged to build next-generation AI assistants and explore the potential and challenges associated with this technology.
Biography
Fanghua Ye is currently pursuing a fully-funded PhD in Computer Science under the supervision of Professor Emine Yilmaz and Professor Jun Wang at University College London (UCL). His research interests span a diverse range of fields, including conversational AI, large language models, natural language processing, information retrieval, federated learning, and social network analysis. His research outcomes have been published in top-tier conferences/journals such as ACL, EMNLP, WWW, NeurIPS, ICDM, CIKM, SIGMOD, TKDD, and TOIS. He has extensive academic and industry experience, having served as a research assistant at the Chinese University of Hong Kong and the National University of Singapore, as an area chair of EMNLP 2023, and as a research intern at Amazon Science and Tencent AI Lab.