# The Bias of Harmful Label Associations in Vision-Language Models

arXiv: https://arxiv.org/abs/2402.07329 · HTML: https://mlsanity.com/paper/2402.07329/

## Key facts

- Cited by 5 papers in the map (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 9 of its 11 references that are in the map.
- First posted on arXiv 11 Feb 2024.

## Placed in

- Multimodal LLMs / vision-language models / [Social bias and fairness in VLMs](https://mlsanity.com/mllm/social-bias-and-fairness-in-vlms/)

## Builds on (most cited first, 9 of 9)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models](https://mlsanity.com/paper/2301.12597/) (2023)
- [Evaluating CLIP: Towards Characterization of Broader Capabilities and Downstream Implications](https://mlsanity.com/paper/2108.02818/) (2021)
- [Towards Measuring Fairness in AI: the Casual Conversations Dataset](https://mlsanity.com/paper/2104.02821/) (2021)
- [Counterfactually Measuring and Eliminating Social Bias in Vision-Language Pre-training Models](https://mlsanity.com/paper/2207.01056/) (2022)
- [Fairness Indicators for Systematic Assessments of Visual Feature Extractors](https://mlsanity.com/paper/2202.07603/) (2022)
- [Model-Agnostic Gender Debiased Image Captioning](https://mlsanity.com/paper/2304.03693/) (2023)
- [Does Progress On Object Recognition Benchmarks Improve Real-World Generalization?](https://mlsanity.com/paper/2307.13136/) (2023)
- [Casual Conversations v2: Designing a large consent-driven dataset to measure algorithmic bias and robustness](https://mlsanity.com/paper/2211.05809/) (2022)

## Built on by (most cited first, 5 of 5)

- [An Introduction to Vision-Language Modeling](https://mlsanity.com/paper/2405.17247/) (2024)
- [Bias in the Picture: Benchmarking VLMs with Social-Cue News Images and LLM-as-Judge Assessment](https://mlsanity.com/paper/2509.19659/) (2025)
- [From Global to Local: Social Bias Transfer in CLIP](https://mlsanity.com/paper/2508.17750/) (2025)
- [Surgeons Are Indian Males and Speech Therapists Are White Females: Auditing Biases in Vision-Language Models for Healthcare Professionals](https://mlsanity.com/paper/2510.06280/) (2025)
- [Guardrail-Agnostic Societal Bias Evaluation in Large Vision-Language Models](https://mlsanity.com/paper/2608.29590/) (2026)

## Abstract (arXiv)

Despite the remarkable performance of foundation vision-language models, the shared representation space for text and vision can also encode harmful label associations detrimental to fairness. While prior work has uncovered bias in vision-language models' (VLMs) classification performance across geography, work has been limited along the important axis of harmful label associations due to a lack of rich, labeled data. In this work, we investigate harmful label associations in the recently released Casual Conversations datasets containing more than 70,000 videos. We study bias in the frequency of harmful label associations across self-provided labels for age, gender, apparent skin tone, and physical adornments across several leading VLMs. We find that VLMs are 4-7x more likely to harmfully classify individuals with darker skin tones. We also find scaling transformer encoder model size leads to higher confidence in harmful predictions. Finally, we find improvements on standard vision tasks across VLMs does not address disparities in harmful label associations.