TopoNC, a topology-aware neural collapse framework for Graph GCD, is proposed and a Topology-Conditioned Pseudo-Labeling strategy that integrates Sinkhorn-based global balancing, Old-Class Top-K Admission Masking, and Neighborhood-Consensus Screening is designed, to reliably guide feature collapse.
Xi Xu, Zhong Zhang, Hong-Liang Wang et al.· Proceedings of the 32nd ACM...· 0 citations
This work finds that finetuning a reward model to guide the policy model is more robust than directly finetuning the policy model, and proposes AgentRM, a generalizable reward model, to guide the policy model for effective test-time search.
Yu Xia, Jing-Ru Fan, Weize Chen et al.· Annual Meeting of the Associ...· 26 citations· ⚡3
Learning a stable yet highly discriminative representation space that can simultaneously recognize known categories and discover novel ones from limited labeled data is fundamental to Generalized Category Discovery (GCD) on graphs. Recently, Neural Collapse (NC) theory has emerged as a powerful geometric principle for...
Xu Xi, Zhong Zhang, Hongliang Wang et al.· Proceedings of the 32nd ACM...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
