Harry Shomer

ladybird.jpg

I’m an Assistant Professor in the Department of Computer Science & Engineering (CSE) at UT Arlington, starting in Fall’25. Before that, I earned my PhD in CSE from Michigan State University in 2025 under Dr. Jiliang Tang and my B.S in CS at CUNY – Brooklyn College in 2019. My work has been published at top conference including NeurIPS, ICLR, KDD, EMNLP, TheWebConf, ACL, and CIKM. I have also received numerous awards including the MSU Engineering Distinguished (EDS) fellowship and the NRT-IMPACTS fellowship.

My research has focused on data mining and machine learning. My main research interest is machine learning on graphs. Some of the topics in this area that I’m currently interested in, include: link prediction, OOD generalization, retrieval-augmented generation on graphs, graph foundation models, and graph generation. I am also interested in Trustworthy AI and more recently the use of AI in Education.

[Recruitment] I am looking to recruit PhD students for Spring’26 and Fall’26. Please see here for more information.

News

Jul 01, 2025 Our paper “Automated Label Placement on Maps via LLMs” is accepted at the AI4DE Workshop at KDD’25!
May 22, 2025 New preprint on incorporating higher-order information for Temporal LP [pdf]
May 16, 2025 Three papers accepted by KDD’25! (1 research track and 2 in D&B)
Apr 17, 2025 📢 We’re organizing a workshop at KDD’25 - Machine Learning on Graphs in the Era of Generative Artificial Intelligence. Papers are due May 8th!
Apr 10, 2025 Paper accepted by EDM’25!
Mar 19, 2025 New preprint - “Empowering GraphRAG with Knowledge Filtering and Integration” [pdf]
Feb 18, 2025 New preprint studying the effectiveness of RAG vs. GraphRAG [pdf]
Jan 05, 2025 New survey on Retrieval-Augmented Generation (RAG) with Graphs [pdf]

Selected Publications

  1. NeurIPS’23
    Evaluating Graph Neural Networks for Link Prediction: Current Pitfalls and New Benchmarking
    Harry Shomer*, Juanhui Li*, Haitao Mao, and 5 more authors
    In Advances in Neural Information Processing Systems, 2023
    PDF
  2. KDD’25 (D&B)
    Understanding the Generalizability of Link Predictors Under Distribution Shifts on Graphs
    Harry Shomer*, Jay Revolinsky*, and Jiliang Tang
    In Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2025
    PDF
  3. KDD’24
    LPFormer: An Adaptive Graph Transformer for Link Prediction
    Harry Shomer, Yao Ma, Haitao Mao, and 3 more authors
    In Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2024
    PDF