[Eecs_phd] EECS Research Seminar Talk with CS Interview Candidate Dr. Juanhui Li-Link Prediction Revisited: From Evaluation Pitfalls to Language Model Synergies

[Hunter, Tiffany]

Title: Link Prediction Revisited: From Evaluation Pitfalls to Language Model Synergies

Abstract: Link prediction is a fundamental task in graph data analysis and plays a crucial role in advancing graph-based applications in real-world machine learning scenarios. However, several challenges hinder progress in this area. Specifically, 1) existing evaluation frameworks are neither unified nor sufficiently rigorous, leading to inconsistent and often suboptimal results, and 2) graph nodes are frequently associated with textual attributes that contain rich semantic information, which is increasingly abundant. Language models have shown remarkable success in processing textual data to capture semantic insights. However, effectively integrating textual information with graph data to enhance real-world applications remains underexplored. In this talk, I will present my research on advancing link prediction from two key perspectives: 1) identifying evaluation pitfalls across various graph types to inspire more advanced methods for link prediction, and 2) leveraging language models in synergy with link prediction techniques to improve a range of real-world applications. Finally, I will present my vision for the future, focusing on the promotion of graph machine learning, the explainability and interpretability of AI, and interdisciplinary research.

Bio:  Juanhui Li is a final-year Ph.D. candidate in the Department of Computer Science and Engineering at Michigan State University. She received her B.Eng. and M.Eng. degrees in computer science from Sun Yat-sen University in 2017 and 2020, respectively. At MSU, she has been working with Prof. Jiliang Tang in the area of graph machine learning, with a particular focus on link prediction. She has also contributed to research in recommendation systems and natural language processing tasks, particularly leveraging large language models. Juanhui has authored numerous high-quality publications in top AI conferences, including NeurIPS, ICLR, KDD, ACL, and SIGIR. Her contributions extend beyond theoretical research to practical advancements. She is an active open-source contributor, maintaining several widely-used public repositories on GitHub. She collaborates closely with industry partners such as SNAP, IBM, Amazon, and The Home Depot. Juanhui is also a recipient of the 2024 ICDM Female Student Shortlist Candidate.

________________________________________________________________________________
Microsoft Teams Need help?<https://aka.ms/JoinTeamsMeeting?omkt=en-US>
Join the meeting now<https://teams.microsoft.com/l/meetup-join/19%3ameeting_ZWYwOTIxN2UtNmQwZS00NjhiLThiNDUtOTQwOGEwOTQxN2Fj%40thread.v2/0?context=%7b%22Tid%22%3a%22f3308007-477c-4a70-8889-34611817c55a%22%2c%22Oid%22%3a%22685c3f4f-29d5-4141-ada5-0fdeab8480e4%22%7d>
Meeting ID: 237 403 666 325
Passcode: hF6Hu6V4
________________________________
Dial in by phone
+1 614-706-6572,,404597818#<tel:+16147066572,,404597818> United States, Columbus
Find a local number<https://dialin.teams.microsoft.com/8f5f7319-0053-4423-a154-4f8b6e7fb7dd?id=404597818>
Phone conference ID: 404 597 818#
For organizers: Meeting options<https://teams.microsoft.com/meetingOptions/?organizerId=685c3f4f-29d5-4141-ada5-0fdeab8480e4&tenantId=f3308007-477c-4a70-8889-34611817c55a&threadId=19_meeting_ZWYwOTIxN2UtNmQwZS00NjhiLThiNDUtOTQwOGEwOTQxN2Fj@thread.v2&messageId=0&language=en-US> | Reset dial-in PIN<https://dialin.teams.microsoft.com/usp/pstnconferencing>
[https://www.ohio.edu/sites/default/files/2018-11/invite_logo_teams.jpg]
If you encounter issues with this meeting, please visit the Help link. If you are not able to resolve the problems, please contact the meeting organizer to let them know you are having difficulty.
Org help<https://www.ohio.edu/oit/services/collaboration/teams/help>
________________________________________________________________________________

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.ohio.edu/pipermail/eecs_phd/attachments/20250214/9397c1d7/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: text/calendar
Size: 10903 bytes
Desc: not available
URL: <http://listserv.ohio.edu/pipermail/eecs_phd/attachments/20250214/9397c1d7/attachment.ics>