mlsanityData to 14 Oct 2026

Revealing and Reducing Gender Biases in Vision and Language Assistants (VLAs)

arXiv 2410.19314 ↗First posted 25 Oct 2024cs.CY cs.CLcited by 15

Not on the map. Why?

Abstract (arXiv)

Pre-trained large language models (LLMs) have been reliably integrated with visual input for multimodal tasks. The widespread adoption of instruction-tuned image-to-text vision-language assistants (VLAs) like LLaVA and InternVL necessitates evaluating gender biases. We study gender bias in 22 popular open-source VLAs with respect to personality traits, skills, and occupations. Our results show that VLAs replicate human biases likely present in the data, such as real-world occupational imbalances. Similarly, they tend to attribute more skills and positive personality traits to women than to men, and we see a consistent tendency to associate negative personality traits with men. To eliminate the gender bias in these models, we find that fine-tuning-based debiasing methods achieve the best trade-off between debiasing and retaining performance on downstream tasks. We argue for pre-deploying gender bias assessment in VLAs and motivate further development of debiasing strategies to ensure equitable societal outcomes. Code is available at https://github.com/ExplainableML/vla-gender-bias.

Key facts. Cited by 15 papers in the map (data through 14 Oct 2026). Builds on 57 of its 88 references that are in the map. First posted on arXiv 25 Oct 2024.

Builds on

57 of 88 references.

References · 57

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2023GPT-4 Technical Reportcited by 18,396
  3. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,380
  4. 2023Visual Instruction Tuningcited by 7,869
  5. 2023BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Modelscited by 6,137
  6. 2023Improved Baselines with Visual Instruction Tuningcited by 4,418
  7. 2022Flamingo: a Visual Language Model for Few-Shot Learningcited by 4,386
  8. 2021The Power of Scale for Parameter-Efficient Prompt Tuningcited by 3,795
  9. 2023Sigmoid Loss for Language Image Pre-Trainingcited by 3,109
  10. 2023MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Modelscited by 2,711
  11. 2023InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Taskscited by 2,565
  12. 2023Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyondcited by 2,224
38 more
  1. 2023MMBench: Is Your Multi-modal Model an All-around Player?cited by 2,141
  2. 2023MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGIcited by 2,080
  3. 2024Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phonecited by 1,917
  4. 2023MME: A Comprehensive Evaluation Benchmark for Multimodal Large Language Modelscited by 1,652
  5. 2022Reproducible scaling laws for contrastive language-image learningcited by 1,173
  6. 2024How Far Are We to GPT-4V? Closing the Gap to Commercial Multimodal Models with Open-Source Suitescited by 1,050
  7. 2024Cambrian-1: A Fully Open, Vision-Centric Exploration of Multimodal LLMscited by 838
  8. 2023Quantifying Language Models' Sensitivity to Spurious Features in Prompt Design or: How I learned to start worrying about prompt formattingcited by 817
  9. 2023Aligning Large Multimodal Models with Factually Augmented RLHFcited by 712
  10. 2018Gender Bias in Coreference Resolution: Evaluation and Debiasing Methodscited by 659
  11. 2023LLM-Pruner: On the Structural Pruning of Large Language Modelscited by 650
  12. 2023Bias and Fairness in Large Language Models: A Surveycited by 639
  13. 2024VLMEvalKit: An Open-Source Toolkit for Evaluating Large Multi-Modality Modelscited by 455
  14. 2024InternLM2 Technical Reportcited by 397
  15. 2019Measuring Bias in Contextualized Word Representationscited by 276
  16. 2022Easily Accessible Text-to-Image Generation Amplifies Demographic Stereotypes at Large Scalecited by 262
  17. 2023Stable Bias: Analyzing Societal Representations in Diffusion Modelscited by 226
  18. 2024MobileVLM V2: Faster and Stronger Baseline for Vision Language Modelcited by 174
  19. 2021Evaluating CLIP: Towards Characterization of Broader Capabilities and Downstream Implicationscited by 128
  20. 2024Efficient Multimodal Learning from Data-centric Perspectivecited by 122
  21. 2022A Prompt Array Keeps the Bias Away: Debiasing Vision-Language Models with Adversarial Learningcited by 104
  22. 2023SEED-Bench-2: Benchmarking Multimodal Large Language Modelscited by 97
  23. 2023DeAR: Debiasing Vision-Language Models with Additive Residualscited by 70
  24. 2023The Bias Amplification Paradox in Text-to-Image Generationcited by 70
  25. 2024Red Teaming Visual Language Modelscited by 65
  26. 2024Examining Gender and Racial Bias in Large Vision-Language Models Using a Novel Dataset of Parallel Imagescited by 55
  27. 2021Worst of Both Worlds: Biases Compound in Pre-trained Vision-and-Language Modelscited by 52
  28. 2023Uncurated Image-Text Datasets: Shedding Light on Demographic Biascited by 52
  29. 2023VisoGender: A dataset for benchmarking gender bias in image-text pronoun resolutioncited by 52
  30. 2021A Step Toward More Inclusive People Annotations for Fairnesscited by 46
  31. 2024Social Bias Evaluation for Large Language Models Requires Prompt Variationscited by 40
  32. 2022Quantifying Social Biases Using Templates is Unreliablecited by 39
  33. 2022Quantifying Societal Bias Amplification in Image Captioningcited by 37
  34. 2024GenderBias-VL: Benchmarking Gender Bias in Vision Language Models via Counterfactual Probingcited by 31
  35. 2022Counterfactually Measuring and Eliminating Social Bias in Vision-Language Pre-training Modelscited by 28
  36. 2024Uncovering Bias in Large Vision-Language Models at Scale with Counterfactualscited by 28
  37. 2023Survey of Social Bias in Vision-Language Modelscited by 26
  38. 2024VLBiasBench: A Comprehensive Benchmark for Evaluating Bias in Large Vision-Language Modelcited by 26

50 most cited of 57 All 57 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 8; 2026: 7.

15 papers.

Sort:

By area:

  1. 2025A Large Scale Analysis of Gender Biases in Text-to-Image Generative Modelscited by 14
  2. 2026Interpretable Debiasing of Vision-Language Models for Social Fairnesscited by 6
  3. 2025Bias in Gender Bias Benchmarks: How Spurious Features Distort Evaluationcited by 5
  4. 2025Person-Centric Annotations of LAION-400M: Auditing Bias and Its Transfer to Modelscited by 4
  5. 2025VIGNETTE: Socially Grounded Bias Evaluation for Vision-Language Modelscited by 3
  6. 2025DSO: Direct Steering Optimization for Bias Mitigationcited by 3
  7. 2025From Individuals to Interactions: Benchmarking Gender Bias in Multimodal Large Language Models from the Lens of Social Relationshipcited by 2
  8. 2025Bias Beyond Demographics: Probing Decision Boundaries in Black-Box LVLMs via Counterfactual VQAcited by 2
  9. 2025Ask Me Again Differently: GRAS for Measuring Bias in Vision Language Models on Gender, Race, Age, and Skin Tonecited by 2
  10. 2026Cultural Counterfactuals: Evaluating Cultural Biases in Large Vision-Language Models with Counterfactual Examplescited by 2
  11. 2026Benchmarking Bias Mitigation Toward Fairness Without Harm from Vision to LVLMscited by 1
  12. 2026Vision-Language Models Suppress Female Representations Under Ambiguous Inputcited by 1
3 more
  1. 2026Guardrail-Agnostic Societal Bias Evaluation in Large Vision-Language Modelscited by 0
  2. 2026Knowing When Not to Answer: Abstention and Refusal Reasoning in Vision--Language Modelscited by 0
  3. 2026ViD: Vision-Dominant Gender Bias Mitigation for Large Vision-Language Modelscited by 0