Learning with noisy labels

Description from the archive archive 2025-07-28.

Benchmarks archive 2025-07-28

20 leaderboard tables shown for this task, 20 with rows (a “benchmark” on this site is a table with at least one row, as on /sota), ordered by row count. “Best model” is the first row in the archive's own order at snapshot; nothing is re-ranked here and metric direction is not recorded in the archive. PwC's Trend sparklines are not in the archive, so that column is omitted. 10 shown of 20 until expanded.

Syntology column: samples harvested from the paper's repositories and executed on synthesized fixtures; “ran” is not a correctness claim and does not order the table. A dash means no Syntology record for that paper, not a recorded non-run. Read from the graph 2026-09-24.

Libraries

Not in the archive: the export carries no per-task library table, so there is nothing to show at snapshot 2025-07-28.

Datasets archive 2025-07-28

16 datasets whose archive record lists this task, ordered by the archive's paper count.

Subtasks archive 2025-07-28

No subtask under this task in the archive's task tree.

Parent tasks archive 2025-07-28

Most implemented papers archive 2025-07-28

30 shown of 143 papers with code (249 tagged with this task in all), ordered by repositories listed in the archive, not by stars (the archive holds no stars, so PwC's “Social” and “Latest” sorts cannot be reproduced). Papers without a page here are shown as plain text.

• Sharpness-Aware Minimization for Efficiently Improving Generalization
  3 Oct 2020 18 repositories listed Syntology ran 8 of 20 samples · 12 unverified · 6 pointer-only (licence)
  In today's heavily overparameterized models, the value of the training loss provides few guarantees on model generalization ability.
• Co-teaching: Robust Training of Deep Neural Networks with Extremely Noisy Labels
  18 Apr 2018 5 repositories listed Syntology ran 7 of 7 samples · 0 unverified · 7 pointer-only (licence)
  Deep learning with noisy labels is practically challenging, as the capacity of deep models is so high that they can totally memorize these noisy labels sooner or later during training.
• Protoformer: Embedding Prototypes for Transformers
  25 Jun 2022 4 repositories listed
  This paper proposes Protoformer, a novel self-learning framework for Transformers that can leverage problematic samples for text classification.
• Learning with Noisy Labels Revisited: A Study Using Real-World Human Annotations
  22 Oct 2021 4 repositories listed
  These observations require us to rethink the treatment of noisy labels, and we hope the availability of these two datasets would facilitate the development and evaluation of future learning with noisy label solutions.
• Open-set Label Noise Can Improve Robustness Against Inherent Label Noise
  21 Jun 2021 4 repositories listed Syntology ran 17 of 29 samples · 12 unverified · 29 pointer-only (licence)
  Learning with noisy labels is a practically challenging problem in weakly supervised learning.
• Normalized Loss Functions for Deep Learning with Noisy Labels
  24 Jun 2020 4 repositories listed Syntology ran 1 of 12 samples · 11 unverified
  However, in practice, simply being robust is not sufficient for a loss function to train accurate DNNs.
• Confident Learning: Estimating Uncertainty in Dataset Labels
  31 Oct 2019 4 repositories listed Syntology ran 4 of 29 samples · 25 unverified · 3 pointer-only (licence)
  Confident learning (CL) is an alternative approach which focuses instead on label quality by characterizing and identifying label errors in datasets, based on the principles of pruning noisy data, counting with…
• Symmetric Cross Entropy for Robust Learning with Noisy Labels
  16 Aug 2019 4 repositories listed
  In this paper, we show that DNN learning with Cross Entropy (CE) exhibits overfitting to noisy labels on some classes ("easy" classes), but more surprisingly, it also suffers from significant under learning on some…
• Generalized Cross Entropy Loss for Training Deep Neural Networks with Noisy Labels
  20 May 2018 4 repositories listed
  Here, we present a theoretically grounded set of noise-robust loss functions that can be seen as a generalization of MAE and CCE.
• Probabilistic End-to-end Noise Correction for Learning with Noisy Labels
  19 Mar 2019 3 repositories listed Syntology ran 2 of 4 samples · 2 unverified · 4 pointer-only (licence)
  Deep learning has achieved excellent performance in various computer vision tasks, but requires a lot of training examples with clean labels.
• How does Disagreement Help Generalization against Label Corruption?
  14 Jan 2019 3 repositories listed
  Learning with noisy labels is one of the hottest problems in weakly-supervised learning.
• Dimensionality-Driven Learning with Noisy Labels
  7 Jun 2018 3 repositories listed
  Datasets with significant proportions of noisy (incorrect) class labels present challenges for training accurate Deep Neural Networks (DNNs).
• Mitigating Label Noise on Graph via Topological Sample Selection
  4 Mar 2024 2 repositories listed Syntology ran 6 of 9 samples · 3 unverified · 8 pointer-only (licence)
  Despite the success of the carefully-annotated benchmarks, the effectiveness of existing graph neural networks (GNNs) can be considerably impaired in practice when the real-world graph data is noisily labeled.
• Learning with Noisy Labels through Learnable Weighting and Centroid Similarity
  16 Mar 2023 2 repositories listed
  We introduce a novel method for training machine learning models in the presence of noisy labels, which are prevalent in domains such as medical diagnosis and autonomous driving and have the potential to degrade a…
• From Noisy Prediction to True Label: Noisy Prediction Calibration via Generative Model
  2 May 2022 2 repositories listed Syntology ran 6 of 13 samples · 7 unverified · 6 pointer-only (licence)
  We suggest a new branch of method, Noisy Prediction Calibration (NPC) in learning with noisy labels.
• Co-Correcting: Noise-tolerant Medical Image Classification via mutual Label Correction
  11 Sep 2021 2 repositories listed
  With the development of deep learning, medical image classification has been significantly improved.
• Clusterability as an Alternative to Anchor Points When Learning with Noisy Labels
  10 Feb 2021 2 repositories listed Syntology ran 0 of 9 samples · 9 unverified
  Nonetheless, finding anchor points remains a non-trivial task, and the estimation accuracy is also often throttled by the number of available anchor points.
• When Optimizing f-divergence is Robust with Label Noise
  7 Nov 2020 2 repositories listed
  We show when maximizing a properly defined f-divergence measure with respect to a classifier's predictions and the supervised labels is robust with label noise.
• Early-Learning Regularization Prevents Memorization of Noisy Labels
  30 Jun 2020 2 repositories listed Syntology ran 1 of 1 samples · 0 unverified
  In contrast with existing approaches, which use the model output during early learning to detect the examples with clean labels, and either ignore or attempt to correct the false labels, we take a different route and…
• Combating noisy labels by agreement: A joint training method with co-regularization
  5 Mar 2020 2 repositories listed
  The state-of-the-art approaches "Decoupling" and "Co-teaching+" claim that the "disagreement" strategy is crucial for alleviating the problem of learning with noisy labels.
• DivideMix: Learning with Noisy Labels as Semi-supervised Learning
  18 Feb 2020 2 repositories listed Syntology ran 0 of 2 samples · 2 unverified · 2 pointer-only (licence)
  Two prominent directions include learning with noisy labels and semi-supervised learning by exploiting unlabeled data.
• Peer Loss Functions: Learning from Noisy Labels without Knowing Noise Rates
  8 Oct 2019 2 repositories listed Syntology ran 4 of 6 samples · 2 unverified
  In this work, we introduce a new family of loss functions that we name as peer loss functions, which enables learning from noisy labels and does not require a priori specification of the noise rates.
• L_DMI: An Information-theoretic Noise-robust Loss Function
  8 Sep 2019 2 repositories listed Syntology ran 4 of 5 samples · 1 unverified · 5 pointer-only (licence)
  \emph{To the best of our knowledge, ℒ_(DMI) is the first loss function that is provably robust to instance-independent label noise, regardless of noise pattern, and it can be applied to any existing classification…
• Making Deep Neural Networks Robust to Label Noise: a Loss Correction Approach
  13 Sep 2016 2 repositories listed Syntology ran 2 of 2 samples · 0 unverified · 2 pointer-only (licence)
  We present a theoretically grounded approach to train deep neural networks, including recurrent networks, subject to class-dependent label noise.
• On the Role of Label Noise in the Feature Learning Process
  25 May 2025 1 repository listed Syntology ran 2 of 8 samples · 6 unverified
  In Stage I, the model perfectly fits all the clean samples (i.
• Detect and Correct: A Selective Noise Correction Method for Learning with Noisy Labels
  19 May 2025 1 repository listed
  Our method identifies potentially noisy samples based on their loss distribution.
• Exploring Video-Based Driver Activity Recognition under Noisy Labels
  16 Apr 2025 1 repository listed
  As an open research topic in the field of deep learning, learning with noisy labels has attracted much attention and grown rapidly over the past ten years.
• Early Stopping Against Label Noise Without Validation Data
  11 Feb 2025 1 repository listed
  Early stopping methods in deep learning face the challenge of balancing the volume of training and validation data, especially in the presence of label noise.
• Learning with Open-world Noisy Data via Class-independent Margin in Dual Representation Space
  19 Jan 2025 1 repository listed
  Learning with Noisy Labels (LNL) aims to improve the model generalization when facing data with noisy labels, and existing methods generally assume that noisy labels come from known classes, called closed-set noise.
• PSSCL: A progressive sample selection framework with contrastive loss designed for noisy labels
  18 Dec 2024 1 repository listed
  Accordingly, their effectiveness is significantly influenced by the precision of the separated clean set, prior knowledge of noise, and the robustness of SSL.

Syntology lines on 15 of the papers shown; no Syntology record for the others (a paper without an arXiv id cannot be joined to the graph, and absence from the graph layer is not a recorded non-run). “Ran” means the sample executed on a synthesized fixture, not that the paper's result was reproduced. Read from the graph 2026-09-24.