CoGe-GCD: Reframing Generalized Category Discovery with Compositional Generalization

• URL: http://arxiv.org/abs/2609.10158v1
• Date: Wed, 09 Sep 2026 13:34:37 GMT
• Title: CoGe-GCD: Reframing Generalized Category Discovery with Compositional Generalization
• Abstract summary: Generalized Category Discovery (GCD) assigns unlabeled instances, mixed with labeled data, to known or novel categories.<n>Existing GCD methods operate on unstructured token features and struggle to extrapolate to novel compositions.<n>We propose CoGe-GCD, which rethinks GCD through compositional generalization with two coupled stages.
• Score: 26.877865526166516
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Generalized Category Discovery (GCD) assigns unlabeled instances, mixed with labeled data, to known or novel categories, requiring human-like compositional reasoning: reusing primitives learned from known classes and deciding when new combinations imply new categories. Existing GCD methods operate on unstructured token features and struggle to extrapolate to novel compositions. We propose CoGe-GCD, which rethinks GCD through compositional generalization with two coupled stages. (i) Compositional Perception structures patch tokens by mapping them to a small vocabulary of primitives and refining token embeddings via competitive token-primitive assignment and information passing, yielding coherent groups for discovery. (ii) Generalizing Induction exploits the induced geometric structure and applies a structure-preserving calibration over spatial relations, maintaining probabilistic semantics while improving extrapolation to unseen primitive combinations. CoGe-GCD is implemented as an inductive-bias module between backbone and projection head, without modifying heads or losses, and can be plugged into diverse GCD frameworks. On standard benchmarks, it consistently improves all-class accuracy, unknown-class number estimation, and geometric quality, with marginal computational overhead. Code is available at https://github.com/lytang63/CoGe-GCD.

Related papers

• CloSeR: Unified Relational Distillation from Closed-Set Teachers for Category Discovery [16.721682363090043]
  CloSeR injects Closed-Set knowledge into Generalized Category Discovery training.<n>Experiments with DINO and DINOv2 backbones on six benchmarks show consistent gains over GCD baselines.
  arXiv  Detail & Related papers  (2026-08-26T12:10:54Z)
• Imputation Meets Clustering: Exploiting Latent Subgroup Structure for Missing Data Recovery [52.29866804948133]
  We propose CAGI (Cluster-Aware Generative Imputation), a framework that reformulates clustering and imputation as a mutually reinforcing co-optimization process.<n>Experiments on 14 benchmark datasets with 15 representative baselines demonstrate the superiority of CAGI.
  arXiv  Detail & Related papers  (2026-07-08T02:50:02Z)
• Learning Like Humans: Analogical Concept Learning for Generalized Category Discovery [48.78384108448773]
  Generalized Category Discovery (GCD) seeks to uncover novel categories in unlabeled data.<n>We introduce the Analogical Textual Concept Generator (ATCG), a plug-and-play module that analogizes from labeled knowledge to new observations.
  arXiv  Detail & Related papers  (2026-03-20T12:59:59Z)
• PartCo: Part-Level Correspondence Priors Enhance Category Discovery [60.37931246658181]
  Generalized Category Discovery (GCD) aims to identify both known and novel categories within unlabeled data.<n>PartCo is a novel framework that enhances category discovery by incorporating part-level visual feature correspondences.<n>Our experiments on multiple benchmark datasets demonstrate that PartCo significantly improves the performance of current GCD approaches.
  arXiv  Detail & Related papers  (2025-09-26T17:59:16Z)
• Consistent Supervised-Unsupervised Alignment for Generalized Category Discovery [49.67913741459179]
  Generalized Category Discovery (GCD) focuses on classifying known categories while simultaneously discovering novel categories from unlabeled data.<n>Previous GCD methods face challenges due to inconsistent optimization objectives and category confusion.<n>We propose the Neural Collapse-inspired Generalized Category Discovery (NC-GCD) framework.
  arXiv  Detail & Related papers  (2025-07-07T07:34:41Z)
• Hyperbolic Category Discovery [12.654574453480635]
  Generalized Category Discovery (GCD) aims to categorize all images in the unlabelled subset, regardless of whether they belong to known or unknown classes.<n>Hyperbolic framework for learning hierarchy-aware representations and classifiers for GCD is presented.
  arXiv  Detail & Related papers  (2025-04-08T15:12:33Z)
• AdaptGCD: Multi-Expert Adapter Tuning for Generalized Category Discovery [15.640670797555579]
  Generalized Category Discovery aims to classify old categories and discover new categories in the unlabeled data.<n>Existing studies on GCD typically devote to transferring the general knowledge from the self-supervised pretrained model to the target GCD task.<n>We propose a novel adapter-tuning-based method named AdaptGCD, which is the first work to introduce the adapter tuning into the GCD task.
  arXiv  Detail & Related papers  (2024-10-29T03:41:47Z)
• Generalized Category Discovery with Clustering Assignment Consistency [56.92546133591019]
  Generalized category discovery (GCD) is a recently proposed open-world task. We propose a co-training-based framework that encourages clustering consistency. Our method achieves state-of-the-art performance on three generic benchmarks and three fine-grained visual recognition datasets.
  arXiv  Detail & Related papers  (2023-10-30T00:32:47Z)
• Dynamic Conceptional Contrastive Learning for Generalized Category Discovery [76.82327473338734]
  Generalized category discovery (GCD) aims to automatically cluster partially labeled data. Unlabeled data contain instances that are not only from known categories of the labeled data but also from novel categories. One effective way for GCD is applying self-supervised learning to learn discriminate representation for unlabeled data. We propose a Dynamic Conceptional Contrastive Learning framework, which can effectively improve clustering accuracy.
  arXiv  Detail & Related papers  (2023-03-30T14:04:39Z)
• Parametric Classification for Generalized Category Discovery: A Baseline Study [70.73212959385387]
  Generalized Category Discovery (GCD) aims to discover novel categories in unlabelled datasets using knowledge learned from labelled samples. We investigate the failure of parametric classifiers, verify the effectiveness of previous design choices when high-quality supervision is available, and identify unreliable pseudo-labels as a key problem. We propose a simple yet effective parametric classification method that benefits from entropy regularisation, achieves state-of-the-art performance on multiple GCD benchmarks and shows strong robustness to unknown class numbers.
  arXiv  Detail & Related papers  (2022-11-21T18:47:11Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.