mlsanityData to 7 Oct 2026

GenderBias-VL: Benchmarking Gender Bias in Vision Language Models via Counterfactual Probing

arXiv 2407.00600 ↗First posted 30 Jun 2024cs.CV cs.AIcited by 31

On the map: Multimodal LLMs / vision-language models → Social bias and fairness in VLMs

Abstract (arXiv)

Large Vision-Language Models (LVLMs) have been widely adopted in various applications; however, they exhibit significant gender biases. Existing benchmarks primarily evaluate gender bias at the demographic group level, neglecting individual fairness, which emphasizes equal treatment of similar individuals. This research gap limits the detection of discriminatory behaviors, as individual fairness offers a more granular examination of biases that group fairness may overlook. For the first time, this paper introduces the GenderBias-VL benchmark to evaluate occupation-related gender bias in LVLMs using counterfactual visual questions under individual fairness criteria. To construct this benchmark, we first utilize text-to-image diffusion models to generate occupation images and their gender counterfactuals. Subsequently, we generate corresponding textual occupation options by identifying stereotyped occupation pairs with high semantic similarity but opposite gender proportions in real-world statistics. This method enables the creation of large-scale visual question counterfactuals to expose biases in LVLMs, applicable in both multimodal and unimodal contexts through modifying gender attributes in specific modalities. Overall, our GenderBias-VL benchmark comprises 34,581 visual question counterfactual pairs, covering 177 occupations. Using our benchmark, we extensively evaluate 15 commonly used open-source LVLMs (eg, LLaVA) and state-of-the-art commercial APIs, including GPT-4o and Gemini-Pro. Our findings reveal widespread gender biases in existing LVLMs. Our benchmark offers: (1) a comprehensive dataset for occupation-related gender bias evaluation; (2) an up-to-date leaderboard on LVLM biases; and (3) a nuanced understanding of the biases presented by these models. footnoteThe dataset and code are available at the hrefhttps://genderbiasvl.github.io/website.

Key facts. Cited by 31 papers in the map (data through 7 Oct 2026). Placed in Social bias and fairness in VLMs, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 80 of its 110 references that are in the map. First posted on arXiv 30 Jun 2024.

Builds on

80 of 110 references.

Closest ancestors · 9

Same direction, newest first

  1. 2024VLBiasBench: A Comprehensive Benchmark for Evaluating Bias in Large Vision-Language Model20 Jun
  2. 2024Uncovering Bias in Large Vision-Language Models at Scale with Counterfactuals30 May
  3. 2024Uncovering Bias in Large Vision-Language Models with Counterfactuals29 Mar
  4. 2024Examining Gender and Racial Bias in Large Vision-Language Models Using a Novel Dataset of Parallel Images8 Feb
  5. 2023SocialCounterfactuals: Probing and Mitigating Intersectional Social Biases in Vision-Language Models with Counterfactual Examples30 Nov
  6. 2023Survey of Social Bias in Vision-Language Models24 Sep
  7. 2023VisoGender: A dataset for benchmarking gender bias in image-text pronoun resolution21 Jun
  8. 2023MultiModal Bias: Introducing a Framework for Stereotypical Bias Assessment beyond Gender and Race in Vision Language Models16 Mar
  9. 2022Counterfactually Measuring and Eliminating Social Bias in Vision-Language Pre-training Models3 Jul

Other references · 71

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2020Language Models are Few-Shot Learnerscited by 32,199
  3. 2023GPT-4 Technical Reportcited by 18,132
  4. 2022Training language models to follow instructions with human feedbackcited by 15,712
  5. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 13,958
  6. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 12,865
  7. 2023Segment Anythingcited by 8,158
  8. 2023Visual Instruction Tuningcited by 7,792
  9. 2023BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Modelscited by 6,069
  10. 2023Improved Baselines with Visual Instruction Tuningcited by 4,362
  11. 2023SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesiscited by 4,290
  12. 2022LAION-5B: An open large-scale dataset for training next generation image-text modelscited by 4,131
38 more
  1. 2023Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detectioncited by 3,274
  2. 2023InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuningcited by 2,855
  3. 2022InstructPix2Pix: Learning to Follow Image Editing Instructionscited by 2,591
  4. 2023Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyondcited by 2,179
  5. 2023MMBench: Is Your Multi-modal Model an All-around Player?cited by 2,105
  6. 2016Equality of Opportunity in Supervised Learningcited by 2,000
  7. 2022Reproducible scaling laws for contrastive language-image learningcited by 1,163
  8. 2023Kosmos-2: Grounding Multimodal Large Language Models to the Worldcited by 1,075
  9. 2016Semantics derived automatically from language corpora contain human-like biasescited by 894
  10. 2023Shikra: Unleashing Multimodal LLM's Referential Dialogue Magiccited by 891
  11. 2017Counterfactual Fairnesscited by 802
  12. 2023EVA-CLIP: Improved Training Techniques for CLIP at Scalecited by 786
  13. 2024Eyes Wide Shut? Exploring the Visual Shortcomings of Multimodal LLMscited by 719
  14. 2014Certifying and removing disparate impactcited by 715
  15. 2023Aligning Large Multimodal Models with Factually Augmented RLHFcited by 703
  16. 2023MiniGPT-v2: large language model as a unified interface for vision-language multi-task learningcited by 617
  17. 2023LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Modelcited by 602
  18. 2023mPLUG-Owl2: Revolutionizing Multi-modal Large Language Model with Modality Collaborationcited by 545
  19. 2020Null It Out: Guarding Protected Attributes by Iterative Nullspace Projectioncited by 442
  20. 2017Men Also Like Shopping: Reducing Gender Bias Amplification using Corpus-level Constraintscited by 420
  21. 2023RLHF-V: Towards Trustworthy MLLMs via Behavior Alignment from Fine-grained Correctional Human Feedbackcited by 404
  22. 2019Bias in Bios: A Case Study of Semantic Representation Bias in a High-Stakes Settingcited by 346
  23. 2021Multimodal datasets: misogyny, pornography, and malignant stereotypescited by 326
  24. 2024InternLM-XComposer2: Mastering Free-form Text-Image Composition and Comprehension in Vision-Language Large Modelcited by 326
  25. 2023InternLM-XComposer: A Vision-Language Large Model for Advanced Text-image Comprehension and Compositioncited by 297
  26. 2023MIMIC-IT: Multi-Modal In-Context Instruction Tuningcited by 264
  27. 2022Easily Accessible Text-to-Image Generation Amplifies Demographic Stereotypes at Large Scalecited by 262
  28. 2017OpenNMT: Open-source Toolkit for Neural Machine Translationcited by 251
  29. 2023Large Language Models Sensitivity to The Order of Options in Multiple-Choice Questionscited by 249
  30. 2018Women also Snowboard: Overcoming Bias in Captioning Modelscited by 236
  31. 2023Stable Bias: Analyzing Societal Representations in Diffusion Modelscited by 225
  32. 2020Towards Debiasing Sentence Representationscited by 170
  33. 2023LAMM: Language-Assisted Multi-Modal Instruction-Tuning Dataset, Framework, and Benchmarkcited by 169
  34. 2021Evaluating CLIP: Towards Characterization of Broader Capabilities and Downstream Implicationscited by 128
  35. 2021Harms of Gender Exclusivity and Challenges in Non-Binary Representation in Language Technologiescited by 128
  36. 2021Dual Attention Suppression Attack: Generate Adversarial Camouflage in Physical Worldcited by 127
  37. 2017Fairness Testing: Testing Software for Discriminationcited by 123
  38. 2021Understanding and Evaluating Racial Biases in Image Captioningcited by 108

50 most cited of 71 All 71 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 2; 2025: 21; 2026: 8.

31 papers.

Sort:

By area:

  1. 2025A Comprehensive Survey in LLM(-Agent) Full Stack Safety: Data, Training and Deploymentcited by 110
  2. 2025Vision Language Models are Biasedcited by 63
  3. 2025HumaniBench: A Human-Centric Framework for Large Multimodal Models Evaluationcited by 19
  4. 2025SecureWebArena: A Holistic Security Evaluation Benchmark for LVLM-based Web Agentscited by 17
  5. 2024Revealing and Reducing Gender Biases in Vision and Language Assistants (VLAs)cited by 15
  6. 2025A Large Scale Analysis of Gender Biases in Text-to-Image Generative Modelscited by 12
  7. 2025Pushing the Limits of Safety: A Technical Report on the ATLAS Challenge 2025cited by 11
  8. 2025BBQ-V: Benchmarking Visual Stereotype Bias in Large Multimodal Modelscited by 10
  9. 2025Detoxifying Large Language Models via Autoregressive Reward Guided Representation Editingcited by 9
  10. 2024Social Debiasing for Fair Multi-modal LLMscited by 8
  11. 2025VLDBench Evaluating Multimodal Disinformation with Regulatory Alignmentcited by 8
  12. 2026CtrlAttack: A Unified Attack on World-Model Control in Diffusion Modelscited by 6
19 more
  1. 2025Vision-Language Models display a strong gender biascited by 5
  2. 2025Bias in Gender Bias Benchmarks: How Spurious Features Distort Evaluationcited by 5
  3. 2025Debiasing CLIP: Interpreting and Correcting Bias in Attention Headscited by 4
  4. 2025SMA: Who Said That? Auditing Membership Leakage in Semi-Black-box RAG Controllingcited by 3
  5. 2025Bias in the Picture: Benchmarking VLMs with Social-Cue News Images and LLM-as-Judge Assessmentcited 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. 2025Freeze and Reveal: Exposing Modality Bias in Vision-Language Modelscited 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
  13. 2025When Cars Have Stereotypes: Auditing Demographic Bias in Objects from Text-to-Image Modelscited by 0
  14. 2025Judging by Appearances? Auditing and Intervening Vision-Language Models for Bail Predictioncited by 0
  15. 2025JMMMU-Pro: Image-based Japanese Multi-discipline Multimodal Understanding Benchmark via Vibe Benchmark Constructioncited by 0
  16. 2026DeepBias: Adaptive In-depth Probing of Social Biases in LVLMscited by 0
  17. 2026Mitigating Bias in Large Vision-Language Models via Counterfactual Ensemble Decodingcited by 0
  18. 2026Guardrail-Agnostic Societal Bias Evaluation in Large Vision-Language Modelscited by 0
  19. 2026ViD: Vision-Dominant Gender Bias Mitigation for Large Vision-Language Modelscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2023Probing Intersectional Biases in Vision-Language Models with Counterfactual Examples
  2. 2024Think Before You Act: A Two-Stage Framework for Mitigating Gender Bias Towards Vision-Language Tasks
  3. 2025BioPro: Towards Difference-Aware Gender Fairness for Vision-Language Models
  4. 2025Ask Me Again Differently: GRAS for Measuring Bias in Vision Language Models on Gender, Race, Age, and Skin Tone
  5. 2024A Unified Framework and Dataset for Assessing Societal Bias in Vision-Language Models
  6. 2026Gender Bias in Vision-Language In-Context Learning
  7. 2025Breaking Language Barriers or Reinforcing Bias? A Study of Gender and Racial Disparities in Multilingual Contrastive Vision Language Models
  8. 2026Measuring Social Bias in Vision-Language Models with Face-Only Counterfactuals from Real Photos