# Class2Simi: A Noise Reduction Perspective on Learning with Noisy Labels

arXiv: https://arxiv.org/abs/2006.07831 · HTML: https://mlsanity.com/paper/2006.07831/

## Key facts

- Cited by 25 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 14 Jun 2020.

## Built on by (most cited first, 25 of 25)

- [Understanding and Improving Early Stopping for Learning with Noisy Labels](https://mlsanity.com/paper/2106.15853/) (2021)
- [Sample Selection with Uncertainty of Losses for Learning with Noisy Labels](https://mlsanity.com/paper/2106.00445/) (2021)
- [Instance-dependent Label-noise Learning under a Structural Causal Model](https://mlsanity.com/paper/2109.02986/) (2021)
- [Mitigating Memorization of Noisy Labels by Clipping the Model Prediction](https://mlsanity.com/paper/2212.04055/) (2022)
- [Noise-robust Graph Learning by Estimating and Leveraging Pairwise Interactions](https://mlsanity.com/paper/2106.07451/) (2021)
- [When Noisy Labels Meet Long Tail Dilemmas: A Representation Calibration Method](https://mlsanity.com/paper/2211.10955/) (2022)
- [Removing Adversarial Noise in Class Activation Feature Space](https://mlsanity.com/paper/2104.09197/) (2021)
- [ReSup: Reliable Label Noise Suppression for Facial Expression Recognition](https://mlsanity.com/paper/2305.17895/) (2023)
- [InstanT: Semi-supervised Learning with Instance-dependent Thresholds](https://mlsanity.com/paper/2310.18910/) (2023)
- [Weakly Supervised AUC Optimization: A Unified Partial AUC Approach](https://mlsanity.com/paper/2305.14258/) (2023)
- [ERASE: Error-Resilient Representation Learning on Graphs for Label Noise Tolerance](https://mlsanity.com/paper/2312.08852/) (2023)
- [Modeling Adversarial Noise for Adversarial Training](https://mlsanity.com/paper/2109.09901/) (2021)
- [Joint Class-Affinity Loss Correction for Robust Medical Image Segmentation with Noisy Labels](https://mlsanity.com/paper/2206.07994/) (2022)
- [Regularly Truncated M-estimators for Learning with Noisy Labels](https://mlsanity.com/paper/2309.00894/) (2023)
- [Tackling Noisy Labels with Network Parameter Additive Decomposition](https://mlsanity.com/paper/2403.13241/) (2024)
- [Instance Correction for Learning with Open-set Noisy Labels](https://mlsanity.com/paper/2106.00455/) (2021)
- [Estimating Noisy Class Posterior with Part-level Labels for Noisy Label Learning](https://mlsanity.com/paper/2405.05714/) (2024)
- [ProSelfLC: Progressive Self Label Correction Towards A Low-Temperature Entropy State](https://mlsanity.com/paper/2207.00118/) (2022)
- [Tackling Instance-Dependent Label Noise with Dynamic Distribution Calibration](https://mlsanity.com/paper/2210.05126/) (2022)
- [Towards Weakly Supervised End-to-end Learning for Long-video Action Recognition](https://mlsanity.com/paper/2311.17118/) (2023)
- [Rep-GLS: Report-Guided Generalized Label Smoothing for Robust Disease Detection](https://mlsanity.com/paper/2508.02495/) (2025)
- [Towards Harnessing Feature Embedding for Robust Learning with Noisy Labels](https://mlsanity.com/paper/2206.13025/) (2022)
- [Harnessing label semantics to extract higher performance under noisy label for Company to Industry matching](https://mlsanity.com/paper/2212.01685/) (2022)
- [Multi-Label Noise Transition Matrix Estimation with Label Correlations: Theory and Algorithm](https://mlsanity.com/paper/2309.12706/) (2023)
- [AsyCo: An Asymmetric Dual-task Co-training Model for Partial-label Learning](https://mlsanity.com/paper/2407.15036/) (2024)

## Abstract (arXiv)

Learning with noisy labels has attracted a lot of attention in recent years, where the mainstream approaches are in pointwise manners. Meanwhile, pairwise manners have shown great potential in supervised metric learning and unsupervised contrastive learning. Thus, a natural question is raised: does learning in a pairwise manner mitigate label noise? To give an affirmative answer, in this paper, we propose a framework called Class2Simi: it transforms data points with noisy class labels to data pairs with noisy similarity labels, where a similarity label denotes whether a pair shares the class label or not. Through this transformation, the reduction of the noise rate is theoretically guaranteed, and hence it is in principle easier to handle noisy similarity labels. Amazingly, DNNs that predict the clean class labels can be trained from noisy data pairs if they are first pretrained from noisy data points. Class2Simi is computationally efficient because not only this transformation is on-the-fly in mini-batches, but also it just changes loss computation on top of model prediction into a pairwise manner. Its effectiveness is verified by extensive experiments.