mlsanityData to 14 Oct 2026

Unlocking the Power of Open Set : A New Perspective for Open-Set Noisy Label Learning

arXiv 2305.04203 ↗First posted 7 May 2023cs.LG cs.CVcited by 8

Not on the map. Why?

Abstract (arXiv)

Learning from noisy data has attracted much attention, where most methods focus on closed-set label noise. However, a more common scenario in the real world is the presence of both open-set and closed-set noise. Existing methods typically identify and handle these two types of label noise separately by designing a specific strategy for each type. However, in many real-world scenarios, it would be challenging to identify open-set examples, especially when the dataset has been severely corrupted. Unlike the previous works, we explore how models behave when faced with open-set examples, and find that a part of open-set examples gradually get integrated into certain known classes, which is beneficial for the separation among known classes. Motivated by the phenomenon, we propose a novel two-step contrastive learning method CECL (Class Expansion Contrastive Learning) which aims to deal with both types of label noise by exploiting the useful information of open-set examples. Specifically, we incorporate some open-set examples into closed-set classes to enhance performance while treating others as delimiters to improve representative ability. Extensive experiments on synthetic and real-world datasets with diverse label noise demonstrate the effectiveness of CECL.

Key facts. Cited by 8 papers in the map (data through 14 Oct 2026). Builds on 48 of its 68 references that are in the map. First posted on arXiv 7 May 2023.

Builds on

48 of 68 references.

References · 48

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,639
  3. 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,513
  4. 2016Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translationcited by 1,475
  5. 2020Understanding Contrastive Representation Learning through Alignment and Uniformity on the Hyperspherecited by 1,405
  6. 2018Co-teaching: Robust Training of Deep Neural Networks with Extremely Noisy Labelscited by 1,059
  7. 2018Generalized Cross Entropy Loss for Training Deep Neural Networks with Noisy Labelscited by 997
  8. 2017A Closer Look at Memorization in Deep Networkscited by 942
  9. 2020DivideMix: Learning with Noisy Labels as Semi-supervised Learningcited by 692
  10. 2013Speech Recognition with Deep Recurrent Neural Networkscited by 669
  11. 2016Making Deep Neural Networks Robust to Label Noise: a Loss Correction Approachcited by 665
  12. 2017MentorNet: Learning Data-Driven Curriculum for Very Deep Neural Networks on Corrupted Labelscited by 653
36 more
  1. 2019Symmetric Cross Entropy for Robust Learning with Noisy Labelscited by 500
  2. 2020Early-Learning Regularization Prevents Memorization of Noisy Labelscited by 416
  3. 2019How does Disagreement Help Generalization against Label Corruption?cited by 390
  4. 2018Learning Models with Uniform Performance via Distributionally Robust Optimizationcited by 335
  5. 2020Combating noisy labels by agreement: A joint training method with co-regularizationcited by 289
  6. 2017WebVision Database: Visual Learning and Understanding from Web Datacited by 262
  7. 2017Decoupling "when to update" from "how to update"cited by 230
  8. 2018Using Trusted Data to Train Deep Networks on Labels Corrupted by Severe Noisecited by 216
  9. 2019Probabilistic End-to-end Noise Correction for Learning with Noisy Labelscited by 210
  10. 2019Are Anchor Points Really Indispensable in Label-Noise Learning?cited by 208
  11. 2017CleanNet: Transfer Learning for Scalable Image Classifier Training with Label Noisecited by 181
  12. 2020Entropy Maximization and Meta Classification for Out-Of-Distribution Detection in Semantic Segmentationcited by 133
  13. 2019Learning From Noisy Labels By Regularized Estimation Of Annotator Confusioncited by 127
  14. 2018Out-of-Distribution Detection Using an Ensemble of Self Supervised Leave-out Classifierscited by 126
  15. 2020Self-Adaptive Training: beyond Empirical Risk Minimizationcited by 126
  16. 2019Deep Self-Learning From Noisy Labelscited by 122
  17. 2022Selective-Supervised Contrastive Learning with Noisy Labelscited by 112
  18. 2022Is Out-of-Distribution Detection Learnable?cited by 109
  19. 2018Iterative Learning with Open-set Noisy Labelscited by 108
  20. 2021Learning with Feature-Dependent Label Noise: A Progressive Approachcited by 102
  21. 2020Multi-Objective Interpolation Training for Robustness to Label Noisecited by 86
  22. 2021Jo-SRC: A Contrastive Approach for Combating Noisy Labelscited by 80
  23. 2021Provable Guarantees for Understanding Out-of-distribution Detectioncited by 78
  24. 2020MoPro: Webly Supervised Learning with Momentum Prototypescited by 66
  25. 2021Open-set Label Noise Can Improve Robustness Against Inherent Label Noisecited by 59
  26. 2021NGC: A Unified Framework for Learning with Open-World Noisy Datacited by 58
  27. 2022ProMix: Combating Label Noise via Maximizing Clean Sample Utilitycited by 48
  28. 2023Model Agnostic Sample Reweighting for Out-of-Distribution Learningcited by 44
  29. 2019IMAE for Noise-Robust Learning: Mean Absolute Error Does Not Treat Examples Equally and Gradient Magnitude's Variance Matterscited by 33
  30. 2021Webly Supervised Fine-Grained Recognition: Benchmark Datasets and An Approachcited by 32
  31. 2022On Learning Contrastive Representations for Learning with Noisy Labelscited by 21
  32. 2020Salvage Reusable Samples from Noisy Data for Robust Learningcited by 19
  33. 2020Suppressing Mislabeled Data via Grouping and Self-Attentioncited by 17
  34. 2020S2OSC: A Holistic Semi-Supervised Approach for Open Set Classificationcited by 15
  35. 2022Regularization Penalty Optimization for Addressing Data Quality Variance in OoD Algorithmscited by 4
  36. 2021Learning from Crowds with Sparse and Imbalanced Annotationscited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 3; 2025: 4; 2026: 1.

8 papers.

  1. 2024Forgetting, Ignorance or Myopia: Revisiting Key Challenges in Online Continual Learningcited by 12
  2. 2024NoisyEQA: Benchmarking Embodied Question Answering Against Noisy Queriescited by 10
  3. 2025Sample Selection via Contrastive Fragmentation for Noisy Label Regressioncited by 4
  4. 2025Open set label noise learning with robust sample selection and margin-guided modulecited by 1
  5. 2025GHOST: Gaussian Hypothesis Open-Set Techniquecited by 1
  6. 2024REPAIR: Rank Correlation and Noisy Pair Half-replacing with Memory for Noisy Correspondencecited by 0
  7. 2025Learning from Uncertain Similarity and Unlabeled Datacited by 0
  8. 2026When Accuracy Is Not Enough: Uncertainty Collapse between Noisy Label Learning and Out-of-Distribution Detectioncited by 0