mlsanityData to 7 Oct 2026

Beyond Hard Labels: Investigating data label distributions

arXiv 2207.06224 ↗First posted 13 Jul 2022cs.CV cs.LGcited by 7

Not on the map. Why?

Abstract (arXiv)

High-quality data is a key aspect of modern machine learning. However, labels generated by humans suffer from issues like label noise and class ambiguities. We raise the question of whether hard labels are sufficient to represent the underlying ground truth distribution in the presence of these inherent imprecision. Therefore, we compare the disparity of learning with hard and soft labels quantitatively and qualitatively for a synthetic and a real-world dataset. We show that the application of soft labels leads to improved performance and yields a more regular structure of the internal feature space.

Key facts. Cited by 7 papers in the map (data through 7 Oct 2026). Builds on 18 of its 28 references that are in the map. First posted on arXiv 13 Jul 2022.

Builds on

18 of 28 references.

References · 18

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 11,979
  3. 2016Densely Connected Convolutional Networkscited by 6,691
  4. 2016SGDR: Stochastic Gradient Descent with Warm Restartscited by 4,993
  5. 2020FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidencecited by 2,080
  6. 2020Big Self-Supervised Models are Strong Semi-Supervised Learnerscited by 1,375
  7. 2020DivideMix: Learning with Noisy Labels as Semi-supervised Learningcited by 691
  8. 2020Learning from Noisy Labels with Deep Neural Networks: A Surveycited by 580
  9. 2017Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning resultscited by 470
  10. 2020Early-Learning Regularization Prevents Memorization of Noisy Labelscited by 414
  11. 2021Pervasive Label Errors in Test Sets Destabilize Machine Learning Benchmarkscited by 388
  12. 2020Are we done with ImageNet?cited by 329
6 more
  1. 2021Learning with Noisy Labels Revisited: A Study Using Real-World Human Annotationscited by 253
  2. 2019Human uncertainty makes classification more robustcited by 231
  3. 2021Re-labeling ImageNet: from Single to Multi-Labels, from Global to Localized Labelscited by 117
  4. 2020Beyond Cats and Dogs: Semi-supervised Classification of fuzzy labels with overclusteringcited by 11
  5. 2021A data-centric approach for improving ambiguous labels with combined semi-supervised classification and clusteringcited by 9
  6. 2021On Data-centric Mythscited by 4

Built on by

Cited by year

Citing papers per year: 2022: 1; 2023: 3; 2024: 2; 2025: 1.

7 papers.

  1. 2024Data-Driven Estimation of the False Positive Rate of the Bayes Binary Classifier via Soft Labelscited by 4
  2. 2024Predict the Next Word: Humans exhibit uncertainty in this task and language models _____cited by 4
  3. 2023Label Smarter, Not Harder: CleverLabel for Faster Annotation of Ambiguous Image Classification with Higher Qualitycited by 3
  4. 2023Evaluating Classification Systems Against Soft Labels with Fuzzy Precision and Recallcited by 3
  5. 2022Opportunistic hip fracture risk prediction in Men from X-ray: Findings from the Osteoporosis in Men (MrOS) Studycited by 2
  6. 2023Annotating Ambiguous Images: General Annotation Strategy for High-Quality Data with Real-World Biomedical Validationcited by 1
  7. 2025Learning from Ambiguous Data with Hard Labelscited by 0