mlsanityData to 14 Oct 2026

Edu-MMBias: A Three-Tier Multimodal Benchmark for Auditing Social Bias in Vision-Language Models under Educational Contexts

arXiv 2604.10200 ↗First posted 11 Apr 2026cs.AI cs.CVcited by 0

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

Abstract (arXiv)

As Vision-Language Models (VLMs) become integral to educational decision-making, ensuring their fairness is paramount. However, current text-centric evaluations neglect the visual modality, leaving an unregulated channel for latent social biases. To bridge this gap, we present Edu-MMBias, a systematic auditing framework grounded in the tri-component model of attitudes from social psychology. This framework diagnoses bias across three hierarchical dimensions: cognitive, affective, and behavioral. Utilizing a specialized generative pipeline that incorporates a self-correct mechanism and human-in-the-loop verification, we synthesize contamination-resistant student profiles to conduct a holistic stress test on state-of-the-art VLMs. Our extensive audit reveals critical, counter-intuitive patterns: models exhibit a compensatory class bias favoring lower-status narratives while simultaneously harboring deep-seated health and racial stereotypes. Crucially, we find that visual inputs act as a safety backdoor, triggering a resurgence of biases that bypass text-based alignment safeguards and revealing a systematic misalignment between latent cognition and final decision-making. The contributions of this paper are available at: https://anonymous.4open.science/r/EduMMBias-63B2.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Social bias and fairness in VLMs, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 14 of its 50 references that are in the map. First posted on arXiv 11 Apr 2026.

Builds on

14 of 50 references.

References · 14

  1. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,626
  2. 2022Self-Instruct: Aligning Language Models with Self-Generated Instructionscited by 2,724
  3. 2022Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answeringcited by 1,919
  4. 2016Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddingscited by 1,312
  5. 2020StereoSet: Measuring stereotypical bias in pretrained language modelscited by 924
  6. 2023Vision-Language Models for Vision Tasks: A Surveycited by 839
  7. 2021Self-Diagnosis and Self-Debiasing: A Proposal for Reducing Corpus-Based Bias in NLPcited by 337
  8. 2018Women also Snowboard: Overcoming Bias in Captioning Modelscited by 236
  9. 2024Yo'LLaVA: Your Personalized Language and Vision Assistantcited by 62
  10. 2024Swift Sampler: Efficient Learning of Sampler by 10 Parameterscited by 25
  11. 2024Multi-Modal Proxy Learning Towards Personalized Visual Multiple Clusteringcited by 21
  12. 2024Customized Multiple Clustering via Multi-Modal Subspace Proxy Learningcited by 12
2 more
  1. 2025ArtMentor: AI-Assisted Evaluation of Artworks to Explore Multimodal Large Language Models Capabilitiescited by 9
  2. 2025LaViC: Adapting Large Vision-Language Models to Visually-Aware Conversational Recommendationcited by 4

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025VIGNETTE: Socially Grounded Bias Evaluation for Vision-Language Models
  2. 2024BiasDora: Exploring Hidden Biased Associations in Vision-Language Models
  3. 2024Uncovering Bias in Large Vision-Language Models at Scale with Counterfactuals
  4. 2025VisBias: Measuring Explicit and Implicit Social Biases in Vision Language Models
  5. 2024A Unified Framework and Dataset for Assessing Societal Bias in Vision-Language Models
  6. 2024Uncovering Bias in Large Vision-Language Models with Counterfactuals
  7. 2025Bias in the Picture: Benchmarking VLMs with Social-Cue News Images and LLM-as-Judge Assessment
  8. 2024Identifying Implicit Social Biases in Vision-Language Models