Portfolio item number 1
Short description of portfolio item number 1
Posts by Collection
portfolio
Portfolio item number 2
Short description of portfolio item number 2
publications
Out-of-Distribution Detection on Graphs: A Survey
Published in arXiv, 2025
For the first time, this paper systematically classifies existing out-of-distribution detection (GOOD) methods, reviews representative studies from 2020 to 2025, provides an in-depth analysis of the core principles and common misconceptions in dealing with the task of GOOD detection, and reveals the main challenges facing the field and future research directions.
Recommended citation: Cai, T.; Jiang, Y.; Liu, Y; Li, M.; Huang, C.; and Pan, S. 2025. Out-of-Distribution Detection on Graphs: A Survey. In arXiv: 2502.08105.
Download Paper
ML-GOOD: Towards Multi-Label Graph Out-Of-Distribution Detection
Published in AAAI, 2025
In this paper, we tackle the emerging issue of multi-label graph out-of-distribution (GOOD) detection. We propose a label-specific energy function that effectively integrates multi-label information to compute GOOD energy scores, supported by theoretical analysis. Through extensive experimentation and discussion, we verify the efficacy of our approach, i.e., ML-GOOD. We aim to inspire further research into OOD detection for multi-label graph classification.
Recommended citation: Cai, T., Jiang, Y., Li, M., Huang, C., Wang, Y., & Huang, Q. (2025). ML-GOOD: Towards Multi-Label Graph Out-Of-Distribution Detection. Proceedings of the AAAI Conference on Artificial Intelligence, 39(15), 15650-15658. https://doi.org/10.1609/aaai.v39i15.33718.
Download Paper
HyperNear: Unnoticeable Node Injection Attacks on Hypergraph Neural Networks
Published in ICML, 2025
In this work, we explore adversarial attacks on hypergraph-based models, revealing their structural vulner abilities. We introduce HyperNear, the first node injection attack framework for HNNs, leveraging homophily constraints for stealth. Experiments demonstrate its high attack efficacy, transferability, and unnoticeability. Our analysis of unnoticeability metrics deepens the understanding of hypergraph structure and model robustness, paving the way for future work on strengthening hypergraph-based models.
Recommended citation: Cai T, Jiang Y, Li M, et al. HyperNear: Unnoticeable Node Injection Attacks on Hypergraph Neural Networks[C]//Forty-second International Conference on Machine Learning.
Download Paper
talks
Paper Title Number 1
Published:
under review
Recommended citation: Your Name, You. (2009). "Paper Title Number 1." Journal 1. 1(1).
Conference Proceeding talk 3 on Relevant Topic in Your Field
Published:
This is a description of your conference proceedings talk, note the different field in type. You can put anything in this field.
teaching
Teaching experience 1
Undergraduate course, University 1, Department, 2014
This is a description of a teaching experience. You can use markdown like any other post.
Teaching experience 2
Workshop, University 1, Department, 2015
This is a description of a teaching experience. You can use markdown like any other post.