Keynote Speaker
August 5 (Wednesday)
Dr. Dilek Hakkani-Tür is a Professor of Computer Science at University of Illinois Urbana-Champaign and an Amazon Scholar (at Amazon Health Science). Recently, she worked as a senior principal scientist at Amazon Alexa AI focusing on enabling natural dialogues with machines (2018-2023). Prior to that, she was a dialogue researcher at Google (2016-2018), a principal researcher at Microsoft Research (2010-2016), International Computer Science Institute (ICSI, 2006-2010) and AT&T Labs-Research (2001-2005). She received her BSc degree from Middle East Technical Univ, in 1994, and MSc and PhD degrees from Bilkent Univ., Department of Computer Engineering, in 1996 and 2000, respectively.
Dr. Hakkani-Tür’s research interests include conversational AI, natural language and speech processing, spoken dialogue systems, and machine learning for language processing. She has over 100 patents that were granted and co-authored more than 300 papers in natural language and speech processing. She received several best paper awards for publications she co-authored on conversational systems, including her earlier work on active learning for dialogue systems, from IEEE Signal Processing Society, ISCA and EURASIP. She served as an associate editor for IEEE Transactions on Audio, Speech and Language Processing (2005-2008), member of the IEEE Speech and Language Technical Committee (2009-2014), area editor for speech and language processing for Elsevier’s Digital Signal Processing Journal and IEEE Signal Processing Letters (2011-2013), and served on the ISCA Advisory Council (2015-2019). She also served as the Editor-in-Chief of the IEEE/ACM Transactions on Audio, Speech and Language Processing (2019-2021), and an IEEE Distinguished Industry Speaker (2021). Currently, she serves as the SigDial President, co-Editor-in-Chief of Transactions of ACL (TACL) and NAACL board member. She is a fellow of the IEEE (2014), ISCA (2014) and ACL (2024).
Towards Safe and Self-Evolving Conversational Agents
Recent advances in large language model (LLM)-based agents have enabled increasingly sophisticated performance on complex tasks. Nevertheless, many such real-world tasks would benefit from human oversight and guidance through conversational interaction. LLMs show performance degradation over multi-turn dialogues, and training with data from human interactions could remedy this. Given the cost of involving real users during training, user simulators have emerged as a practical surrogate; however, they introduce their own limitations. Compounding these challenges, safety vulnerabilities tend to amplify in multi-turn conversational settings, posing additional risks for deployed systems. In this talk, I will present an overview of our recent studies that aim to address these challenges, with the broader goal of advancing the development of more capable, reliable, and self-improving AI agents.