mlsanityData to 14 Oct 2026

ReLKD: Inter-Class Relation Learning with Knowledge Distillation for Generalized Category Discovery

arXiv 2512.07229 ↗First posted 8 Dec 2025cs.CVcited by 0

On the map: Knowledge distillation → General knowledge distillation methods

Abstract (arXiv)

Generalized Category Discovery (GCD) faces the challenge of categorizing unlabeled data containing both known and novel classes, given only labels for known classes. Previous studies often treat each class independently, neglecting the inherent inter-class relations. Obtaining such inter-class relations directly presents a significant challenge in real-world scenarios. To address this issue, we propose ReLKD, an end-to-end framework that effectively exploits implicit inter-class relations and leverages this knowledge to enhance the classification of novel classes. ReLKD comprises three key modules: a target-grained module for learning discriminative representations, a coarse-grained module for capturing hierarchical class relations, and a distillation module for transferring knowledge from the coarse-grained module to refine the target-grained module's representation learning. Extensive experiments on four datasets demonstrate the effectiveness of ReLKD, particularly in scenarios with limited labeled data. The code for ReLKD is available at https://github.com/ZhouF-ECNU/ReLKD.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in General knowledge distillation methods, one of 12 directions in Knowledge distillation. Builds on 40 of its 48 references that are in the map. First posted on arXiv 8 Dec 2025.

Builds on

40 of 48 references.

Closest ancestors · 2

Same direction, newest first

  1. 2024Rethinking Centered Kernel Alignment in Knowledge Distillation22 Jan
  2. 2023Cluster-aware Semi-supervised Learning: Relational Knowledge Distillation Provably Learns Clustering20 Jul

Other references · 38

  1. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
  2. 2015Distilling the Knowledge in a Neural Networkcited by 10,324
  3. 2020Supervised Contrastive Learningcited by 2,829
  4. 2020FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidencecited by 2,085
  5. 2013Fine-Grained Visual Classification of Aircraftcited by 1,726
  6. 2019Self-training with Noisy Student improves ImageNet classificationcited by 1,311
  7. 2016Paying More Attention to Attention: Improving the Performance of Convolutional Neural Networks via Attention Transfercited by 1,077
  8. 2019Relational Knowledge Distillationcited by 889
  9. 2017Deep Mutual Learningcited by 628
  10. 2019Be Your Own Teacher: Improve the Performance of Convolutional Neural Networks via Self Distillationcited by 497
  11. 2019S4L: Self-Supervised Semi-Supervised Learningcited by 335
  12. 2018Paraphrasing Complex Network: Network Compression via Factor Transfercited by 260
26 more
  1. 2018Knowledge Transfer via Distillation of Activation Boundaries Formed by Hidden Neuronscited by 238
  2. 2019Learning to Discover Novel Visual Categories via Deep Transfer Clusteringcited by 225
  3. 2022Generalized Category Discoverycited by 211
  4. 2022MaskCLIP: Masked Self-Distillation Advances Contrastive Language-Image Pretrainingcited by 187
  5. 2021Open-World Semi-Supervised Learningcited by 165
  6. 2017Learning to cluster in order to transfer across domains and taskscited by 128
  7. 2021Neighborhood Contrastive Learning for Novel Class Discoverycited by 121
  8. 2022Parametric Classification for Generalized Category Discovery: A Baseline Studycited by 113
  9. 2019Multi-class Classification without Multi-class Labelscited by 100
  10. 2023Dynamic Conceptional Contrastive Learning for Generalized Category Discoverycited by 92
  11. 2022PromptCAL: Contrastive Affinity Learning via Auxiliary Prompts for Generalized Novel Category Discoverycited by 87
  12. 2023Learning Semi-supervised Gaussian Mixture Models for Generalized Category Discoverycited by 72
  13. 2024SPTNet: An Efficient Alternative Framework for Generalized Category Discovery with Spatial Prompt Tuningcited by 51
  14. 2023MetaGCD: Learning to Continually Learn in Generalized Category Discoverycited by 42
  15. 2022OpenLDN: Learning to Discover Novel Classes for Open-World Semi-Supervised Learningcited by 37
  16. 2021Meta Discovery: Learning to Discover Novel Classes given Very Limited Datacited by 31
  17. 2024Active Generalized Category Discoverycited by 31
  18. 2025Solving the Catastrophic Forgetting Problem in Generalized Category Discoverycited by 26
  19. 2023Proxy Anchor-based Unsupervised Learning for Continuous Generalized Category Discoverycited by 16
  20. 2023Open-world Semi-supervised Novel Class Discoverycited by 15
  21. 2023Generalized Category Discovery with Large Language Models in the Loopcited by 15
  22. 2024Effective Prompt Pool Learning for Continual Category Discoverycited by 14
  23. 2022Fine-grained Category Discovery under Coarse-grained supervision with Hierarchical Weighted Self-contrastive Learningcited by 13
  24. 2023Towards Distribution-Agnostic Generalized Category Discoverycited by 12
  25. 2024CDAD-Net: Bridging Domain Gaps in Generalized Category Discoverycited by 12
  26. 2024Online Continuous Generalized Category Discoverycited by 5

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026CloSeR: Unified Relational Distillation from Closed-Set Teachers for Category Discovery
  2. 2023Class-relation Knowledge Distillation for Novel Class Discovery
  3. 2025Prior-Constrained Association Learning for Fine-Grained Generalized Category Discovery
  4. 2025GLEAN: Active Generalized Category Discovery with Diverse LLM Feedback
  5. 2022Class-incremental Novel Class Discovery
  6. 2024NC-NCD: Novel Class Discovery for Node Classification
  7. 2025DebGCD: Debiased Learning with Distribution Guidance for Generalized Category Discovery
  8. 2023CORSD: Class-Oriented Relational Self Distillation