# Exposing and Correcting the Gender Bias in Image Captioning Datasets and Models

arXiv: https://arxiv.org/abs/1912.00578 · HTML: https://mlsanity.com/paper/1912.00578/

## Key facts

- Cited by 36 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 2 Dec 2019.

## Built on by (most cited first, 30 of 36)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [DALL-Eval: Probing the Reasoning Skills and Social Biases of Text-to-Image Generation Models](https://mlsanity.com/paper/2202.04053/) (2022)
- [Understanding and Evaluating Racial Biases in Image Captioning](https://mlsanity.com/paper/2106.08503/) (2021)
- [Vision Models Are More Robust And Fair When Pretrained On Uncurated Images Without Supervision](https://mlsanity.com/paper/2202.08360/) (2022)
- [Theories of "Gender" in NLP Bias Research](https://mlsanity.com/paper/2205.02526/) (2022)
- [Examining Gender and Racial Bias in Large Vision-Language Models Using a Novel Dataset of Parallel Images](https://mlsanity.com/paper/2402.05779/) (2024)
- [Visual Haystacks: A Vision-Centric Needle-In-A-Haystack Benchmark](https://mlsanity.com/paper/2407.13766/) (2024)
- [Worst of Both Worlds: Biases Compound in Pre-trained Vision-and-Language Models](https://mlsanity.com/paper/2104.08666/) (2021)
- [Uncurated Image-Text Datasets: Shedding Light on Demographic Bias](https://mlsanity.com/paper/2304.02828/) (2023)
- [MAP: Multimodal Uncertainty-Aware Vision-Language Pre-training Model](https://mlsanity.com/paper/2210.05335/) (2022)
- [Gender Artifacts in Visual Datasets](https://mlsanity.com/paper/2206.09191/) (2022)
- [Measuring Representational Harms in Image Captioning](https://mlsanity.com/paper/2206.07173/) (2022)
- [Generate, but Verify: Reducing Hallucination in Vision-Language Models with Retrospective Resampling](https://mlsanity.com/paper/2504.13169/) (2025)
- [GenderBias-VL: Benchmarking Gender Bias in Vision Language Models via Counterfactual Probing](https://mlsanity.com/paper/2407.00600/) (2024)
- [Underspecification in Scene Description-to-Depiction Tasks](https://mlsanity.com/paper/2210.05815/) (2022)
- [Identifying Implicit Social Biases in Vision-Language Models](https://mlsanity.com/paper/2411.00997/) (2024)
- [Survey of Social Bias in Vision-Language Models](https://mlsanity.com/paper/2309.14381/) (2023)
- [Fairness Indicators for Systematic Assessments of Visual Feature Extractors](https://mlsanity.com/paper/2202.07603/) (2022)
- [MultiModal Bias: Introducing a Framework for Stereotypical Bias Assessment beyond Gender and Race in Vision Language Models](https://mlsanity.com/paper/2303.12734/) (2023)
- [Model-Agnostic Gender Debiased Image Captioning](https://mlsanity.com/paper/2304.03693/) (2023)
- [BendVLM: Test-Time Debiasing of Vision-Language Embeddings](https://mlsanity.com/paper/2411.04420/) (2024)
- [DiffusionPID: Interpreting Diffusion via Partial Information Decomposition](https://mlsanity.com/paper/2406.05191/) (2024)
- [Fairness and Bias Mitigation in Computer Vision: A Survey](https://mlsanity.com/paper/2408.02464/) (2024)
- [A Unified Framework and Dataset for Assessing Societal Bias in Vision-Language Models](https://mlsanity.com/paper/2402.13636/) (2024)
- [Adventures of Trustworthy Vision-Language Models: A Survey](https://mlsanity.com/paper/2312.04231/) (2023)
- [Analyzing Quality, Bias, and Performance in Text-to-Image Generative Models](https://mlsanity.com/paper/2407.00138/) (2024)
- [What's in a Caption? Dataset-Specific Linguistic Diversity and Its Effect on Visual Description Models and Metrics](https://mlsanity.com/paper/2205.06253/) (2022)
- [Distribution Aware Metrics for Conditional Natural Language Generation](https://mlsanity.com/paper/2209.07518/) (2022)
- [Run Like a Girl! Sports-Related Gender Bias in Language and Vision](https://mlsanity.com/paper/2305.14468/) (2023)
- [Evaluating Bias and Fairness in Gender-Neutral Pretrained Vision-and-Language Models](https://mlsanity.com/paper/2310.17530/) (2023)

## Abstract (arXiv)

The task of image captioning implicitly involves gender identification. However, due to the gender bias in data, gender identification by an image captioning model suffers. Also, the gender-activity bias, owing to the word-by-word prediction, influences other words in the caption prediction, resulting in the well-known problem of label bias. In this work, we investigate gender bias in the COCO captioning dataset and show that it engenders not only from the statistical distribution of genders with contexts but also from the flawed annotation by the human annotators. We look at the issues created by this bias in the trained models. We propose a technique to get rid of the bias by splitting the task into 2 subtasks: gender-neutral image captioning and gender classification. By this decoupling, the gender-context influence can be eradicated. We train the gender-neutral image captioning model, which gives comparable results to a gendered model even when evaluating against a dataset that possesses a similar bias as the training data. Interestingly, the predictions by this model on images with no humans, are also visibly different from the one trained on gendered captions. We train gender classifiers using the available bounding box and mask-based annotations for the person in the image. This allows us to get rid of the context and focus on the person to predict the gender. By substituting the genders into the gender-neutral captions, we get the final gendered predictions. Our predictions achieve similar performance to a model trained with gender, and at the same time are devoid of gender bias. Finally, our main result is that on an anti-stereotypical dataset, our model outperforms a popular image captioning model which is trained with gender.