Keynote Speaker
ChengXiang Zhai, UIUC (USA)
Towards a Unified Game-Theoretic Framework for eCommerce Search and Recommendation
Abstract: A general goal in eCommerce (eCom) is to connect users with the right products or services at the right time. Such a connection has been done primarily via either a search system or a recommender system depending on whether a user or a system initiates the process of connection. Despite the close relationship between eCom search and recommendation, research so far on them has not been as integrated as it could be. In this talk, I will present a general game-theoretic framework that eanbles unification of conversational recommender systems and search engines in the same theoretical framework, thus facilitating the fusion of techniques developed separately for recommender systems and search engines and enabling development of a general intelligent eCom system that can support not only search and recommendation but also many other useful interactive functions to optimize user-system collaboration. Using the framework as a roadmap, I will discuss the major opportunities and challenges for future research in eCom, especially those involving the use of large language models such as ChatGPT.
Speaker Bio
ChengXiang Zhai is a Donald Biggar Willett Professor in Engineering of Department of Computer Science at the University of Illinois at Urbana-Champaign, where he also holds a joint appointment at the Carl R. Woese Institute for Genomic Biology, Department of Statistics, and the School of Information Sciences. His research interests are in the general area of intelligent information systems, including specifically intelligent information retrieval, data mining, natural language processing, machine learning, and their applications in domains such as eCommerce, biomedical informatics, and intelligent education systems. He has published over 300 papers in these areas and holds 6 patents. He offers two MOOCs on Coursera and has published a textbook on Text Data Management and Analysis. He served as Associate Editors for major journals in multiple areas (ACM TOIS, ACM TKDD, ACM TIST, IPM,and BMC MIDM), Program Co-Chairs of CIKM'04, NAACL HLT'07, SIGIR'09, and WWW'15, and Conference Co-Chairs of CIKM'16, WSDM'18, and IEEE BigData'20. He is an ACM Fellow and a member of ACM SIGIR Academy. He received numerous awards, including ACM SIGIR Gerard Salton Award, ACM SIGIR Test of Time Paper Award (three times), the US Presidential Early Career Award for Scientists and Engineers (PECASE), Alfred P. Sloan Research Fellowship, IBM Faculty Award, HP Innovation Research Award, Google Research Award, Microsoft Beyond Search Research Award, UIUC Rose Award for Teaching Excellence, and UIUC Campus Award for Excellence in Graduate Student Mentoring. He has graduated 38 PhD students and over 50 master students.
Invited Speakers
Tejaswi Tenneti, Instacart (USA)
Application of LLMs to improve Grocery search
Abstract: This talk will focus on the role of LLMs in enhancing search capabilities of a grocery e-commerce platform, to deliver a more inspiring shopping journey. We will explore how we can use LLMs to suggest and interpret natural language queries, providing accurate and contextually relevant results, enabling users to try new ideas and discover new products in a more efficient manner. The talk will also focus on common challenges faced in integrating LLMs into the search stack, the complexity involved in evaluation of LLM responses particularly from the lens of customer satisfaction.
Speaker Bio
Tejaswi Tenneti Tejaswi Tenneti is currently a Director of Machine Learning at Instacart, the north american leader in online grocery. Prior to Instacart, Tejaswi was a tech lead in machine learning teams at Apple and Oracle where he worked on various applications related to Search and Recommendations for local maps data and Enterprise. Tejaswi holds a BS from IIIT, Allahabad and an MS from Stanford University specializing in AI.
Daniel Campos, Snowflake (USA)
Benchmarking End To End Product Retrieval
Abstract: Effective product retrieval lies at the core of successful online shopping experiences, enabling users to find relevant products efficiently. However, evaluating and comparing the performance of different retrieval systems is challenging due to the lack of standardized evaluation metrics and datasets that accurately represent real-world eCommerce scenarios. Seeking to fix this, we are hosting the 2023 Product Search Track at NIST's TREC Conference. In this talk, we will discuss our dataset, evaluation methods, and what broad research questions we seek to answer.
Speaker Bio
Daniel Campos is a Research Scientist at Snowflake where he works on conversational agents. His Ph.D. is from the University of Illinois Urbana-Champaign, where he focused on building efficient and robust language models for language understanding and generation. His Masters is from the University of Washington, where he focused on curriculum learning for language models. His bachelor’s is from RPI where he worked on traditional NLP and semantic search. Daniel is passionate about making AI more accessible, efficient, and robust. He has experience in developing and deploying state-of-the-art models for various tasks such as sentiment analysis, entity extraction, question answering, and document ranking.