Abstract:In this study, we tackle Generalized Category Discovery (GCD) via a Relational Retrieval perspective, explicitly coupling labeled and unlabeled data through bidirectional knowledge transfer. While existing methods treat these sources separately, missing valuable interaction opportunities, we propose Relational Pattern Consistency (RPC) that enables mutual enhancement. RPC employs One-vs-All classifiers for soft ID/OOD decomposition, then introduces two mechanisms: (i) for known-class preservation, we transfer semantic behavioral alignment; (ii) for category discovery, we leverage the insight that samples from the same category maintain invariant relationships with known-class prototypes, transforming unreliable pseudo-labeling into well-defined relational pattern matching. This bidirectional design allows labeled data to guide unlabeled learning while discovering novel categories through their collective relational signatures. Extensive experiments demonstrate RPC achieves state-of-the-art performance on both generic and fine-grained benchmarks.
| Comments: | Accepted by ICMR 2026. Generalized category discovery, semi-supervised learning, contrastive learning |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Multimedia (cs.MM) |
| Cite as: | arXiv:2605.09420 [cs.CV] |
| (or arXiv:2605.09420v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2605.09420 arXiv-issued DOI via DataCite (pending registration) |
|
| Related DOI: | https://doi.org/10.1145/3805622.3810732
DOI(s) linking to related resources |
Submission history
From: Chunqi Guo [view email]
[v1]
Sun, 10 May 2026 08:46:59 UTC (684 KB)