mlsanityData to 14 Oct 2026

Consistent Supervised-Unsupervised Alignment for Generalized Category Discovery

arXiv 2507.04725 ↗First posted 7 Jul 2025cs.CVcited by 6

Not on the map. Why?

Abstract (arXiv)

Generalized Category Discovery (GCD) focuses on classifying known categories while simultaneously discovering novel categories from unlabeled data. However, previous GCD methods face challenges due to inconsistent optimization objectives and category confusion. This leads to feature overlap and ultimately hinders performance on novel categories. To address these issues, we propose the Neural Collapse-inspired Generalized Category Discovery (NC-GCD) framework. By pre-assigning and fixing Equiangular Tight Frame (ETF) prototypes, our method ensures an optimal geometric structure and a consistent optimization objective for both known and novel categories. We introduce a Consistent ETF Alignment Loss that unifies supervised and unsupervised ETF alignment and enhances category separability. Additionally, a Semantic Consistency Matcher (SCM) is designed to maintain stable and consistent label assignments across clustering iterations. Our method achieves strong performance on multiple GCD benchmarks, significantly enhancing novel category accuracy and demonstrating its effectiveness.

Key facts. Cited by 6 papers in the map (data through 14 Oct 2026). Builds on 42 of its 51 references that are in the map. First posted on arXiv 7 Jul 2025.

Builds on

42 of 51 references.

References · 42

  1. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  2. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,968
  3. 2016Gaussian Error Linear Units (GELUs)cited by 3,502
  4. 2013Fine-Grained Visual Classification of Aircraftcited by 1,726
  5. 2019Contrastive Representation Distillationcited by 748
  6. 2020Prevalence of Neural Collapse during the terminal phase of deep learning trainingcited by 719
  7. 2021Open-Set Recognition: a Good Closed-Set Classifier is All You Need?cited by 353
  8. 2021Towards Open World Object Detectioncited by 297
  9. 2014Towards Open World Recognitioncited by 226
  10. 2022Generalized Category Discoverycited by 211
  11. 2021Exploring Deep Neural Networks via Layer-Peeled Model: Minority Collapse in Imbalanced Trainingcited by 166
  12. 2021Open-World Semi-Supervised Learningcited by 165
30 more
  1. 2021A Unified Objective for Novel Class Discoverycited by 162
  2. 2021Neural Collapse Under MSE Loss: Proximity to and Dynamics on the Central Pathcited by 158
  3. 2020Automatically Discovering and Learning New Visual Categories with Ranking Statisticscited by 143
  4. 2020Neural collapse with unconstrained featurescited by 142
  5. 2022Forward Compatible Few-Shot Class-Incremental Learningcited by 127
  6. 2023Neural Collapse Inspired Feature-Classifier Alignment for Few-Shot Class Incremental Learningcited by 116
  7. 2022Parametric Classification for Generalized Category Discovery: A Baseline Studycited by 113
  8. 2022Inducing Neural Collapse in Imbalanced Learning: Do We Really Need a Learnable Classifier at the End of Deep Neural Network?cited by 100
  9. 2021An Unconstrained Layer-Peeled Perspective on Neural Collapsecited by 87
  10. 2022PromptCAL: Contrastive Affinity Learning via Auxiliary Prompts for Generalized Novel Category Discoverycited by 87
  11. 2023Learning Semi-supervised Gaussian Mixture Models for Generalized Category Discoverycited by 72
  12. 2021Dissecting Supervised Contrastive Learningcited by 64
  13. 2019The Herbarium Challenge 2019 Datasetcited by 60
  14. 2024SPTNet: An Efficient Alternative Framework for Generalized Category Discovery with Spatial Prompt Tuningcited by 51
  15. 2023Understanding Imbalanced Semantic Segmentation Through Neural Collapsecited by 46
  16. 2022Parametric Information Maximization for Generalized Category Discoverycited by 39
  17. 2023Generalized Neural Collapse for a Large Number of Classescited by 39
  18. 2024Contrastive Mean-Shift Learning for Generalized Category Discoverycited by 39
  19. 2020Explicit regularization and implicit bias in deep network classifiers trained with the square losscited by 37
  20. 2024SelEx: Self-Expertise in Fine-Grained Generalized Category Discoverycited by 34
  21. 2024Active Generalized Category Discoverycited by 31
  22. 2025ProtoGCD: Unified and Unbiased Prototype Learning for Generalized Category Discoverycited by 19
  23. 2023Proxy Anchor-based Unsupervised Learning for Continuous Generalized Category Discoverycited by 16
  24. 2024Open World Object Detection: A Surveycited by 13
  25. 2025DualCP: Rehearsal-Free Domain-Incremental Learning via Dual-Level Concept Prototypecited by 7
  26. 2025Generalized Category Discovery via Reciprocal Learning and Class-Wise Distribution Regularizationcited by 6
  27. 2023Category Adaptation Meets Projected Distillation in Generalized Continual Category Discoverycited by 5
  28. 2024Online Continuous Generalized Category Discoverycited by 5
  29. 2023Neural Collapse Inspired Federated Learning with Non-iid Datacited by 4
  30. 2025Learn by Reasoning: Analogical Weight Generation for Few-Shot Class-Incremental Learningcited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 2; 2026: 4.

6 papers.

  1. 2025PartCo: Part-Level Correspondence Priors Enhance Category Discoverycited by 2
  2. 2025Generalized Fine-Grained Category Discovery with Multi-Granularity Conceptual Expertscited by 2
  3. 2026GOAL: Geometrically Optimal Alignment for Continual Generalized Category Discoverycited by 1
  4. 2026Learning Like Humans: Analogical Concept Learning for Generalized Category Discoverycited by 0
  5. 2026Fourier Self-Supervision for Fine-Grained Generalized Category Discoverycited by 0
  6. 2026BPG: Balancing Plasticity and Generalization for Domain Incremental Learningcited by 0