Learning with Noisy Labels

Overview

Learning with noisy labels asks how models can learn useful decision rules when training targets disagree with the underlying truth, as often happens with crowd annotations, medical labels, and automatically generated supervision. Common approaches include robust losses and noise-transition models, sample selection and reweighting, co-teaching, label correction, and self-supervised or representation-based methods that identify suspicious examples without trusting their labels. Methods are tested on controlled synthetic corruption and real-world benchmarks such as CIFAR and Clothing1M, as well as in medical imaging, graphs, and other domains where noise may depend on the example or annotator. A central challenge is that noise structure and data difficulty matter: small-loss filtering can discard hard but correct examples, and success on closed-set synthetic benchmarks may not transfer to open-set noise or deployment settings without clean labels for evaluation.

Latest papers

CardsList

168 papers17 papers in the last four weeks, up 89% on the four weeks before. 0.1% of all new papers.

  1. Robust Flow Matching under Target Corruption and Label Noise

    Nov 9, 2026Mert Can Kurucu, Erik Englesson, Hossein AzizpourFlow MatchingLearning with Noisy Labels

  2. Dual-Granularity Learning for Regression with Continuous Noisy Labels

    Nov 7, 2026Hui Guo, Boyu Wang, Grace Y. YiNoisy-Label LearningLearning with Noisy Labels

  3. On the Reliability of Majority-Vote Pseudo-Labels in Test-Time Reinforcement Learning

    Nov 6, 2026Haoyu Wang, Ru Wang, Yuanyuan Wang +2Test-Time RLRL for Language Model Reasoning

  4. Leakage- and Label-Aware Benchmarking of Class-Imbalance Strategies for Multimodal Chest X-Ray and EHR Classification

    Nov 5, 2026Dang-Khoa Bui, Quang-Huy Tran, Minh-Khoi Nguyen-Bui +4Medical Image BenchmarksData Leakage

  5. An Information Budget for Label-Noise Memorization, and What Statistical Complexity Tracks: A Study with Multilayer Perceptrons

    Nov 3, 2026Nancy Baygorrea C., Otávio da Fonseca Martins GomesNeural Network MemorizationLearning with Noisy Labels

  6. Peripheral prototypical contrastive loss for supervised medical image segmentation

    Nov 1, 2026Takuro Shimaya, Ryosaku Shino, Masahiro SaikoPrototypical Contrastive LearningMedical Image Segmentation

  7. Reduced Cost Influence Functions for Predict-then-Optimize under Noisy Data

    Nov 1, 2026William Zhang, Saurabh Amin, Georgia PerakisPredict-Then-OptimizeInfluence Functions

  8. Robust Reward Machine Learning and Execution in Noisy Environments

    Oct 31, 2026Roko Parac, Lorenzo Nodari, Daniel Furelos-Blanco +3Non-Markovian RLLearning with Noisy Labels

  9. Reliability-Weighted Direct Preference Optimization: Penalized Weights for Robust Preference Learning

    Oct 30, 2026Bin Luo, Param Venkat Vivek Kesireddy, Anil VallepuDirect Preference OptimizationPreference Learning

  10. When is early stopping beneficial under label noise? A phase diagram for kernel gradient flow

    Oct 24, 2026Fabrizio Boncoraglio, Lenka ZdeborováEarly StoppingLearning with Noisy Labels

  11. Improved automatic speech-based cognitive assessment using neuropsychologically calibrated diagnostic labels

    Oct 20, 2026Madhurananda Pahar, Caitlin Illingworth, Adam Hampshire +5Speech-Based Dementia DetectionClinical Speech Processing

  12. Understanding Data Augmentation under Label Noise

    Oct 20, 2026Hong Zheng, Fei TengData AugmentationLearning with Noisy Labels

  13. CAVA: Complementary-Label-Aware Data Valuation

    Oct 19, 2026Tan-Ha Mai, Hsuan-Tien Lin, Vu NguyenData ValuationLearning with Noisy Labels

  14. Hybrid supervised instance segmentation by learning label noise suppression

    Oct 14, 2026Linwei Chen, Ying Fu, Shaodi You +1Instance SegmentationPseudo-Labeling

  15. Noisy-Label Calibration with Multiple Unreliable Labels

    Oct 14, 2026Keisuke Hanada, Futoshi FutamiModel CalibrationProbability Calibration

  16. Learning What to Trust in Multimodal Learning under Noisy Supervision

    Oct 8, 2026Jiashuo Zou, Xiaobo XiaLearning with Noisy LabelsMultimodal Classification

  17. Optimal Transport Reweighting for Robust Learning under Spurious Correlations and Label Noise

    Oct 1, 2026Sung Ho Jo, Seonghwi Kim, Wonsang Yun +1Distribution Shift RobustnessSpurious Correlation Robustness

  18. When an Evaluation Rule Writes Training Labels: Measuring Human-Reference Forgiveness in NAVSIM

    Sep 27, 2026Jiaxuan Guo, Jingxin Yang, Jiaqi Ye +4Autonomous Driving BenchmarksLearning with Noisy Labels

  19. Adaptive Ensemble Selection for Noisy Labels on Tabular Data

    Sep 26, 2026Faizaan Ali, Inwon Kang, Oshani SeneviratneLearning with Noisy LabelsEnsemble Learning

  20. Different Corruptions, Different Signals: Uncertainty and Loss in Federated Data Quality

    Sep 25, 2026Bradley Scott, Zeqi Luo, Edmond S. L. HoFederated LearningUncertainty Quantification

  21. GERIS: A Game-Theoretic Framework for Filtering Instance-Dependent Label Noise in License Plate Data Augmentation

    Sep 22, 2026Seyedeh Sara Jalili Shani, Rouhollah Ahmadian, Amin Rahmani +2Game-Theoretic RLData Selection

  22. Efficient Robust Learning at the Information-Theoretic Limit

    Sep 15, 2026Adam R. Klivans, Konstantinos Stavropoulos, Sergei Tikhonov +1Boolean Function LearningLearning with Noisy Labels

  23. From Phase Transition to Systemic Failure: A Decoupled Analytics Framework for GNN Robustness

    Sep 15, 2026Shuai Yan, Dan Peng, Jie Li +1Graph Neural NetworksDistribution Shift Robustness