paper-with-me

홈 › Papers

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

2025-01-28 · Sarah Bonna, Yu-Cheng Huang, Ekaterina Novozhilova, Sejin Paik, Zhengyang Shan, Michelle Yilin Feng, Ge Gao, Yonish Tayal, Rushil Kulkarni, Jialin Yu, Nupur Divekar, Deepti Ghadiyaram, Derry Wijaya, Margrit Betke

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.

📄 PDF Abstract BibTeX arXiv:2501.18642

Code (0)

등록된 구현이 없습니다.

Tasks

AttributeImage Generation

Similar Papers 제목 키워드 기반

FairSteer: Inference Time Debiasing for LLMs with Dynamic Activation Steering

2025-04-20 · Yichen Li, Zhiting Fan, Ruizhe Chen, Xiaotang Gai 외

Large language models (LLMs) are prone to capturing biases from training corpus, leading to potential negative social impacts. Existing prompt-based debiasing methods exhibit instability due to their sensitivity to promp…

counterfactualFairnessQuestion AnsweringText Generation

Steering LLMs Towards Unbiased Responses: A Causality-Guided Debiasing Framework

2024-03-13 · Jingling Li, Zeyu Tang, Xiaoyu Liu, Peter Spirtes 외

Large language models (LLMs) can easily generate biased and discriminative responses. As LLMs tap into consequential decision-making (e.g., hiring and healthcare), it is of crucial importance to develop strategies to mit…

Decision Making

ADEPT: A DEbiasing PrompT Framework

2022-11-10 · Ke Yang, Charles Yu, Yi Fung, Manling Li 외

Several works have proven that finetuning is an applicable approach for debiasing contextualized word embeddings. Similarly, discrete prompts with semantic meanings have shown to be effective in debiasing tasks. With unf…

AttributeLanguage ModellingWord Embeddings

Inference-Time Selective Debiasing

2024-07-27 · Gleb Kuzmin, Neemesh Yadav, Ivan Smirnov, Timothy Baldwin 외

We propose selective debiasing -- an inference-time safety mechanism that aims to increase the overall quality of models in terms of prediction performance and fairness in the situation when re-training a model is prohib…

Fairnesstext-classificationText Classification

GGSS: Geodesic-Gated Spherical Steering for Inference-Time Debiasing of Generative Vision-Language Models

2026-08-26 · Yiqun Sun, Junyu Chen, Pengfei Wei, Lawrence B. Hsieh arxiv

Generative vision-language models (VLMs) are increasingly used in human-centered settings, yet they can produce demographically biased outputs even when images differ only in controlled attributes such as perceived race …