mlsanityData to 7 Oct 2026

Distribution-Balanced Loss for Multi-Label Classification in Long-Tailed Datasets

arXiv 2007.09654 ↗First posted 19 Jul 2020cs.CV cs.LGcited by 122

Not on the map. Why?

Abstract (arXiv)

We present a new loss function called Distribution-Balanced Loss for the multi-label recognition problems that exhibit long-tailed class distributions. Compared to conventional single-label classification problem, multi-label recognition problems are often more challenging due to two significant issues, namely the co-occurrence of labels and the dominance of negative labels (when treated as multiple binary classification problems). The Distribution-Balanced Loss tackles these issues through two key modifications to the standard binary cross-entropy loss: 1) a new way to re-balance the weights that takes into account the impact caused by label co-occurrence, and 2) a negative tolerant regularization to mitigate the over-suppression of negative labels. Experiments on both Pascal VOC and COCO show that the models trained with this new loss function achieve significant performance gains over existing methods. Code and models are available at: https://github.com/wutong16/DistributionBalancedLoss .

Key facts. Cited by 122 papers in the map (data through 7 Oct 2026). First posted on arXiv 19 Jul 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 27; 2022: 26; 2023: 31; 2024: 20; 2025: 12; 2026: 6.

122 papers.

Sort:

By area:

  1. 2021Deep Long-Tailed Learning: A Surveycited by 416
  2. 2021Query2Label: A Simple Transformer Way to Multi-Label Classificationcited by 126
  3. 2021MetaSAug: Meta Semantic Augmentation for Long-Tailed Visual Recognitioncited by 99
  4. 2021ML-Decoder: Scalable and Versatile Classification Headcited by 78
  5. 2022A Survey on Long-Tailed Visual Recognitioncited by 75
  6. 2021Adaptive Methods for Real-World Domain Generalizationcited by 66
  7. 2021Trustworthy Long-Tailed Classificationcited by 51
  8. 2021Adversarial Robustness under Long-Tailed Distributioncited by 50
  9. 2022Large Loss Matters in Weakly Supervised Multi-Label Classificationcited by 38
  10. 2021VideoLT: Large-scale Long-tailed Video Recognitioncited by 35
  11. 2021Robust Long-Tailed Learning under Label Noisecited by 35
  12. 2021Balancing Methods for Multi-label Text Classification with Long-Tailed Class Distributioncited by 34
38 more
  1. 2022Learning Imbalanced Data with Vision Transformerscited by 34
  2. 2025MAT-Agent: Adaptive Multi-Agent Training Optimizationcited by 34
  3. 2023FoodLMM: A Versatile Food Assistant using Large Multi-modal Modelcited by 32
  4. 2021Class-Balanced Distillation for Long-Tailed Visual Recognitioncited by 27
  5. 2021Relational Subsets Knowledge Distillation for Long-tailed Retinal Diseases Recognitioncited by 27
  6. 2021MovieCuts: A New Dataset and Benchmark for Cut Type Recognitioncited by 26
  7. 2021Multi-label Classification with Partial Annotations using Class-aware Selective Losscited by 25
  8. 2022Delving into Semantic Scale Imbalancecited by 22
  9. 2022Open Long-Tailed Recognition in a Dynamic Worldcited by 21
  10. 2022A Survey of Methods for Addressing Class Imbalance in Deep-Learning Based Natural Language Processingcited by 20
  11. 2023TagCLIP: A Local-to-Global Framework to Enhance Open-Vocabulary Multi-Label Classification of CLIP Without Trainingcited by 19
  12. 2024SFC: Shared Feature Calibration in Weakly Supervised Semantic Segmentationcited by 19
  13. 2022Relieving Long-tailed Instance Segmentation via Pairwise Class Balancecited by 18
  14. 2023Multi-label affordance mapping from egocentric visioncited by 18
  15. 2021Sewer-ML: A Multi-Label Sewer Defect Classification Dataset and Benchmarkcited by 17
  16. 2023Learning in Imperfect Environment: Multi-Label Classification with Long-Tailed Distribution and Partial Labelscited by 17
  17. 2023Fed-GraB: Federated Long-tailed Learning with Self-Adjusting Gradient Balancercited by 17
  18. 2024End-to-End Neuro-Symbolic Reinforcement Learning with Textual Explanationscited by 17
  19. 2023LMPT: Prompt Tuning with Class-Specific Embedding Loss for Long-tailed Multi-Label Visual Recognitioncited by 15
  20. 2024Long-Tailed Anomaly Detection with Learnable Class Namescited by 14
  21. 2022Optimizing Class Distribution in Memory for Multi-Label Online Continual Learningcited by 12
  22. 2023CheXFusion: Effective Fusion of Multi-View Features using Transformers for Long-Tailed Chest X-Ray Classificationcited by 12
  23. 2023Action-slot: Visual Action-centric Representations for Multi-label Atomic Activity Recognition in Traffic Scenescited by 12
  24. 2024Beyond Pixels: Semi-Supervised Semantic Segmentation with a Multi-scale Patch-based Multi-Label Classifiercited by 12
  25. 2021GCNBoost: Artwork Classification by Label Propagation through a Knowledge Graphcited by 11
  26. 2021Predicate correlation learning for scene graph generationcited by 11
  27. 2021Training Over-parameterized Models with Non-decomposable Objectivescited by 11
  28. 2022RankSeg: Adaptive Pixel Classification with Image Category Ranking for Segmentationcited by 11
  29. 2023Toward Robustness in Multi-label Classification: A Data Augmentation Strategy against Imbalance and Noisecited by 11
  30. 2021PLM: Partial Label Masking for Imbalanced Multi-label Classificationcited by 10
  31. 2021Hierarchical Knowledge Guided Learning for Real-world Retinal Diseases Recognitioncited by 10
  32. 2022Facial Action Unit Recognition With Multi-models Ensemblingcited by 10
  33. 2022ELM: Embedding and Logit Margins for Long-Tail Learningcited by 10
  34. 2024On the selection and effectiveness of pseudo-absences for species distribution modeling with deep learningcited by 10
  35. 2025Long-tailed Recognition with Model Rebalancingcited by 10
  36. 2022Chairs Can be Stood on: Overcoming Object Bias in Human-Object Interaction Detectioncited by 9
  37. 2022MultiGuard: Provably Robust Multi-label Classification against Adversarial Examplescited by 9
  38. 2024CLIP-Decoder : ZeroShot Multilabel Classification using Multimodal CLIP Aligned Representationcited by 9

50 most cited of 122 All 122 (JSON)