# DebiasPI: Inference-time Debiasing by Prompt Iteration of a Text-to-Image Generative Model

arXiv: https://arxiv.org/abs/2501.18642 · HTML: https://mlsanity.com/paper/2501.18642/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 27 references that are in the map.
- First posted on arXiv 28 Jan 2025.

## Builds on (most cited first, 12 of 12)

- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2307.15217/) (2023)
- [Easily Accessible Text-to-Image Generation Amplifies Demographic Stereotypes at Large Scale](https://mlsanity.com/paper/2211.03759/) (2022)
- [DALL-Eval: Probing the Reasoning Skills and Social Biases of Text-to-Image Generation Models](https://mlsanity.com/paper/2202.04053/) (2022)
- [General Facial Representation Learning in a Visual-Linguistic Manner](https://mlsanity.com/paper/2112.03109/) (2021)
- [How well can Text-to-Image Generative Models understand Ethical Natural Language Interventions?](https://mlsanity.com/paper/2210.15230/) (2022)
- [Evaluating the Social Impact of Generative AI Systems in Systems and Society](https://mlsanity.com/paper/2306.05949/) (2023)
- [Ethical-Advice Taker: Do Language Models Understand Natural Language Interventions?](https://mlsanity.com/paper/2106.01465/) (2021)
- ['Person' == Light-skinned, Western Man, and Sexualization of Women of Color: Stereotypes in Stable Diffusion](https://mlsanity.com/paper/2310.19981/) (2023)
- [Biases in Generative Art -- A Causal Look from the Lens of Art History](https://mlsanity.com/paper/2010.13266/) (2020)
- [Smiling Women Pitching Down: Auditing Representational and Presentational Gender Biases in Image Generative AI](https://mlsanity.com/paper/2305.10566/) (2023)

## Built on by (most cited first, 3 of 3)

- [Training-Free Debiasing of Diffusion Models via CLIP-Guided Denoising Optimization](https://mlsanity.com/paper/2607.00817/) (2026)
- [Embedding Arithmetic: A Lightweight, Tuning-Free Framework for Post-hoc Bias Mitigation in Text-to-Image Models](https://mlsanity.com/paper/2604.18167/) (2026)
- [KG-FairDiff: Knowledge Graph-Guided Prompt Refinement for Demographically Fair Text-to-Image Generation](https://mlsanity.com/paper/2606.01282/) (2026)

## Abstract (arXiv)

Ethical intervention prompting has emerged as a tool to counter demographic biases of text-to-image generative AI models. Existing solutions either require to retrain the model or struggle to generate images that reflect desired distributions on gender and race. We propose an inference-time process called DebiasPI for Debiasing-by-Prompt-Iteration that provides prompt intervention by enabling the user to control the distributions of individuals' demographic attributes in image generation. DebiasPI keeps track of which attributes have been generated either by probing the internal state of the model or by using external attribute classifiers. Its control loop guides the text-to-image model to select not yet sufficiently represented attributes, With DebiasPI, we were able to create images with equal representations of race and gender that visualize challenging concepts of news headlines. We also experimented with the attributes age, body type, profession, and skin tone, and measured how attributes change when our intervention prompt targets the distribution of an unrelated attribute type. We found, for example, if the text-to-image model is asked to balance racial representation, gender representation improves but the skin tone becomes less diverse. Attempts to cover a wide range of skin colors with various intervention prompts showed that the model struggles to generate the palest skin tones. We conducted various ablation studies, in which we removed DebiasPI's attribute control, that reveal the model's propensity to generate young, male characters. It sometimes visualized career success by generating two-panel images with a pre-success dark-skinned person becoming light-skinned with success, or switching gender from pre-success female to post-success male, thus further motivating ethical intervention prompting with DebiasPI.