# Mitigating Label Bias via Decoupled Confident Learning

arXiv: https://arxiv.org/abs/2307.08945 · HTML: https://mlsanity.com/paper/2307.08945/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 42 references that are in the map.
- First posted on arXiv 18 Jul 2023.

## Builds on (most cited first, 12 of 12)

- [Equality of Opportunity in Supervised Learning](https://mlsanity.com/paper/1610.02413/) (2016)
- [Co-teaching: Robust Training of Deep Neural Networks with Extremely Noisy Labels](https://mlsanity.com/paper/1804.06872/) (2018)
- [Confident Learning: Estimating Uncertainty in Dataset Labels](https://mlsanity.com/paper/1911.00068/) (2019)
- [Dealing with Disagreements: Looking Beyond the Majority Vote in Subjective Annotations](https://mlsanity.com/paper/2110.05719/) (2021)
- [Normalized Loss Functions for Deep Learning with Noisy Labels](https://mlsanity.com/paper/2006.13554/) (2020)
- [Measuring a hate speech spectrum with faceted Rasch item response theory and perspective-aware, explainable-by-design deep learning](https://mlsanity.com/paper/2009.10277/) (2020)
- [Fairness Evaluation in Presence of Biased Noisy Labels](https://mlsanity.com/paper/2003.13808/) (2020)
- [Algorithmic Fairness in Business Analytics: Directions for Research and Practice](https://mlsanity.com/paper/2207.10991/) (2022)
- [Leveraging Expert Consistency to Improve Algorithmic Decision Support](https://mlsanity.com/paper/2101.09648/) (2021)
- [The effect of differential victim crime reporting on predictive policing systems](https://mlsanity.com/paper/2102.00128/) (2021)
- [Same Same, But Different: Conditional Multi-Task Learning for Demographic-Specific Toxicity Detection](https://mlsanity.com/paper/2302.07372/) (2023)
- [More Data Can Lead Us Astray: Active Data Acquisition in the Presence of Label Bias](https://mlsanity.com/paper/2207.07723/) (2022)

## Built on by (most cited first, 1 of 1)

- [Towards Fairness under Label Bias in Image Segmentation: Impact, Measurement and Mitigation](https://mlsanity.com/paper/2605.06891/) (2026)

## Abstract (arXiv)

Growing concerns regarding algorithmic fairness have led to a surge in methodologies to mitigate algorithmic bias. However, such methodologies largely assume that observed labels in training data are correct. This is problematic because bias in labels is pervasive across important domains, including healthcare, hiring, and content moderation. In particular, human-generated labels are prone to encoding societal biases. While the presence of labeling bias has been discussed conceptually, there is a lack of methodologies to address this problem. We propose a pruning method -- Decoupled Confident Learning (DeCoLe) -- specifically designed to mitigate label bias. After illustrating its performance on a synthetic dataset, we apply DeCoLe in the context of hate speech detection, where label bias has been recognized as an important challenge, and show that it successfully identifies biased labels and outperforms competing approaches.