# Social Debiasing for Fair Multi-modal LLMs

arXiv: https://arxiv.org/abs/2408.06569 · HTML: https://mlsanity.com/paper/2408.06569/

## Key facts

- Cited by 8 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 55 of its 75 references that are in the map.
- First posted on arXiv 13 Aug 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, 30 of 55)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models](https://mlsanity.com/paper/2301.12597/) (2023)
- [Generative Adversarial Networks](https://mlsanity.com/paper/2203.00667/) (2022)
- [mixup: Beyond Empirical Risk Minimization](https://mlsanity.com/paper/1710.09412/) (2017)
- [Improved Baselines with Visual Instruction Tuning](https://mlsanity.com/paper/2310.03744/) (2023)
- [Flamingo: a Visual Language Model for Few-Shot Learning](https://mlsanity.com/paper/2204.14198/) (2022)
- [SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis](https://mlsanity.com/paper/2307.01952/) (2023)
- [BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation](https://mlsanity.com/paper/2201.12086/) (2022)
- [InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning](https://mlsanity.com/paper/2305.06500/) (2023)
- [Making the V in VQA Matter: Elevating the Role of Image Understanding in Visual Question Answering](https://mlsanity.com/paper/1612.00837/) (2016)
- [Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond](https://mlsanity.com/paper/2308.12966/) (2023)
- [MMBench: Is Your Multi-modal Model an All-around Player?](https://mlsanity.com/paper/2307.06281/) (2023)
- [Evaluating Object Hallucination in Large Vision-Language Models](https://mlsanity.com/paper/2305.10355/) (2023)
- [Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings](https://mlsanity.com/paper/1607.06520/) (2016)
- [VL-BERT: Pre-training of Generic Visual-Linguistic Representations](https://mlsanity.com/paper/1908.08530/) (2019)
- [PaLI: A Jointly-Scaled Multilingual Language-Image Model](https://mlsanity.com/paper/2209.06794/) (2022)
- [DeepSeek-VL: Towards Real-World Vision-Language Understanding](https://mlsanity.com/paper/2403.05525/) (2024)
- [Bias and Fairness in Large Language Models: A Survey](https://mlsanity.com/paper/2309.00770/) (2023)
- [Language Is Not All You Need: Aligning Perception with Language Models](https://mlsanity.com/paper/2302.14045/) (2023)
- [TextCaps: a Dataset for Image Captioning with Reading Comprehension](https://mlsanity.com/paper/2003.12462/) (2020)
- [Learning Adversarially Fair and Transferable Representations](https://mlsanity.com/paper/1802.06309/) (2018)
- [Multimodal datasets: misogyny, pornography, and malignant stereotypes](https://mlsanity.com/paper/2110.01963/) (2021)
- [Word Embeddings Quantify 100 Years of Gender and Ethnic Stereotypes](https://mlsanity.com/paper/1711.08412/) (2017)
- [Debiasing Vision-Language Models via Biased Prompts](https://mlsanity.com/paper/2302.00070/) (2023)
- [Black is to Criminal as Caucasian is to Police: Detecting and Removing Multiclass Bias in Word Embeddings](https://mlsanity.com/paper/1904.04047/) (2019)

## Built on by (most cited first, 8 of 8)

- [Multimodal Large Language Models for Medicine: A Comprehensive Survey](https://mlsanity.com/paper/2504.21051/) (2025)
- [FMBench: Benchmarking Fairness in Multimodal Large Language Models on Medical Tasks](https://mlsanity.com/paper/2410.01089/) (2024)
- [Fair Deepfake Detectors Can Generalize](https://mlsanity.com/paper/2507.02645/) (2025)
- [Measuring Social Bias in Vision-Language Models with Face-Only Counterfactuals from Real Photos](https://mlsanity.com/paper/2601.06931/) (2026)
- [Greater accessibility can amplify discrimination in generative AI](https://mlsanity.com/paper/2603.22260/) (2026)
- [Demographic and Linguistic Bias Evaluation in Omnimodal Language Models](https://mlsanity.com/paper/2604.10014/) (2026)
- [GGSS: Geodesic-Gated Spherical Steering for Inference-Time Debiasing of Generative Vision-Language Models](https://mlsanity.com/paper/2608.25375/) (2026)
- [IMPLICIT-Bench: Measuring Implicit Bias in Text-to-Image Models under Neutral Prompts](https://mlsanity.com/paper/2609.24228/) (2026)

## Abstract (arXiv)

Multi-modal Large Language Models (MLLMs) have dramatically advanced the research field and delivered powerful vision-language understanding capabilities. However, these models often inherit deep-rooted social biases from their training data, leading to uncomfortable responses with respect to attributes such as race and gender. This paper addresses the issue of social biases in MLLMs by i) introducing a comprehensive counterfactual dataset with multiple social concepts (CMSC), which complements existing datasets by providing 18 diverse and balanced social concepts; and ii) proposing a counter-stereotype debiasing (CSD) strategy that mitigates social biases in MLLMs by leveraging the opposites of prevalent stereotypes. CSD incorporates both a novel bias-aware data sampling method and a loss rescaling method, enabling the model to effectively reduce biases. We conduct extensive experiments with four prevalent MLLM architectures. The results demonstrate the advantage of the CMSC dataset and the edge of CSD strategy in reducing social biases compared to existing competing methods, without compromising the overall performance on general multi-modal reasoning benchmarks.