Leveraging Large Language Models to Enhance Comprehensive Graph Learning Capabilities

Speaker: Jiang Zhuoren, Zhejiang University, Department of Information Resources Management, "Hundred‑Talents Program" researcher.

Bio: Jiang is a doctoral supervisor who has published over 60 high‑quality academic papers in top international and domestic venues, leads more than ten projects funded by the National Natural Science Foundation, Ministry of Science and Technology, and other major programs, and serves on several AI and information retrieval committees. He previously advised Alibaba DAMO Academy’s Language Technology Lab and has won the 2013 ACM/IEEE‑CS Joint Conference on Digital Libraries Best Poster Award as well as multiple international AI and data‑algorithm competition championships. His research interests include computational social science, natural language processing, and information retrieval.

Talk Title: Leveraging Large Language Models to Enhance Comprehensive Graph Learning Capabilities

Outline: Although large language models (LLMs) have demonstrated strong pure‑text reasoning abilities, their potential for graph learning remains under‑explored. This presentation first reviews existing technical applications of LLMs on graphs, then surveys literature from the perspectives of different graph‑learning scenarios and the various roles LLMs can play. The analysis reveals that LLMs offer opportunities for a unified, cross‑domain, cross‑task learning framework on graphs. Finally, the talk summarizes potential research directions in this rapidly evolving field.

Audience Benefits:

Gain an overview of the current research status of graph learning adapted to large language models.

Learn how LLMs can facilitate cross‑domain data sources and heterogeneous learning tasks in graph learning.

Identify promising research avenues within this emerging area.