mlsanityData to 14 Oct 2026

MAVias: Mitigate any Visual Bias

arXiv 2412.06632 ↗First posted 9 Dec 2024cs.CVcited by 5

Not on the map. Why?

Abstract (arXiv)

Mitigating biases in computer vision models is an essential step towards the trustworthiness of artificial intelligence models. Existing bias mitigation methods focus on a small set of predefined biases, limiting their applicability in visual datasets where multiple, possibly unknown biases exist. To address this limitation, we introduce MAVias, an open-set bias mitigation approach leveraging foundation models to discover spurious associations between visual attributes and target classes. MAVias first captures a wide variety of visual features in natural language via a foundation image tagging model, and then leverages a large language model to select those visual features defining the target class, resulting in a set of language-coded potential visual biases. We then translate this set of potential biases into vision-language embeddings and introduce an in-processing bias mitigation approach to prevent the model from encoding information related to them. Our experiments on diverse datasets, including CelebA, Waterbirds, ImageNet, and UrbanCars, show that MAVias effectively detects and mitigates a wide range of biases in visual recognition tasks outperforming current state-of-the-art.

Key facts. Cited by 5 papers in the map (data through 14 Oct 2026). Builds on 37 of its 53 references that are in the map. First posted on arXiv 9 Dec 2024.

Builds on

37 of 53 references.

References · 37

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  3. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  4. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  5. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  6. 2024The Llama 3 Herd of Modelscited by 14,715
  7. 2023Mistral 7Bcited by 3,164
  8. 2023Sigmoid Loss for Language Image Pre-Trainingcited by 3,109
  9. 2016Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotationscited by 2,656
  10. 2024Gemma 2: Improving Open Language Models at a Practical Sizecited by 1,998
  11. 2019A Survey on Bias and Fairness in Machine Learningcited by 1,917
  12. 2019Distributionally Robust Neural Networks for Group Shifts: On the Importance of Regularization for Worst-Case Generalizationcited by 1,247
25 more
  1. 2024Qwen2.5-Coder Technical Reportcited by 1,203
  2. 2021Just Train Twice: Improving Group Robustness without Training Group Informationcited by 523
  3. 2020Noise or Signal: The Role of Image Backgrounds in Object Recognitioncited by 343
  4. 2019Don't Take the Easy Way Out: Ensemble Based Methods for Avoiding Known Dataset Biasescited by 280
  5. 2020Gradient Starvation: A Learning Proclivity in Neural Networkscited by 267
  6. 2018Learning Not to Learn: Training Deep Neural Networks with Biased Datacited by 255
  7. 2019Learning De-biased Representations with Biased Representationscited by 199
  8. 2023Bias Runs Deep: Implicit Reasoning Biases in Persona-Assigned LLMscited by 199
  9. 2020REVISE: A Tool for Measuring and Mitigating Bias in Visual Datasetscited by 136
  10. 2021EnD: Entangling and Disentangling deep representations for bias correctioncited by 105
  11. 2022A Whac-A-Mole Dilemma: Shortcuts Come in Multiples Where Mitigating One Amplifies Otherscited by 85
  12. 2021A Survey on Bias in Visual Datasetscited by 84
  13. 2018Robot Learning in Homes: Improving Generalization and Reducing Dataset Biascited by 75
  14. 2023Discover and Cure: Concept-aware Mitigation of Spurious Correlationcited by 71
  15. 2024OpenBias: Open-set Bias Detection in Text-to-Image Generative Modelscited by 61
  16. 2023Discovering and Mitigating Visual Biases through Keyword Explanationcited by 58
  17. 2022Discover and Mitigate Unknown Biases with Debiasing Alternate Networkscited by 47
  18. 2023Identifying Spurious Biases Early in Training through the Lens of Simplicity Biascited by 44
  19. 2022Unbiased Supervised Contrastive Learningcited by 40
  20. 2023Bounds on Representation-Induced Confounding Bias for Treatment Effect Estimationcited by 20
  21. 2023FLAC: Fairness-Aware Representation Learning by Suppressing Attribute-Class Associationscited by 18
  22. 2023Towards Fair Face Verification: An In-depth Analysis of Demographic Biasescited by 14
  23. 2024BAdd: Bias Mitigation through Bias Additioncited by 5
  24. 2024Say My Name: a Model's Bias Discovery Frameworkcited by 4
  25. 2024Be Aware of the Neighborhood Effect: Modeling Selection Bias under Interferencecited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 3; 2026: 2.

5 papers.

  1. 2025Processing and acquisition traces in visual encoders: What does CLIP know about your camera?cited by 3
  2. 2025VB-Mitigator: An Open-source Framework for Evaluating and Advancing Visual Bias Mitigationcited by 2
  3. 2025Did Models Learn Sufficiently? Attribution-Guided Training via Subset-Selected Counterfactual Augmentationcited by 1
  4. 2026Face Age Verification Vulnerabilities Under Simple Appearance Manipulationscited by 0
  5. 2026Invisible Shortcuts: Why Vision Encoders Know Your Cameracited by 0