Stay Fair! Ensuring Group Fairness in Diffusion Models Across Guidance Scales

• URL: http://arxiv.org/abs/2605.28036v1
• Date: Wed, 27 May 2026 06:39:25 GMT
• Title: Stay Fair! Ensuring Group Fairness in Diffusion Models Across Guidance Scales
• Abstract summary: Diffusion models steer conditional generation with a tunable guidance scale to trade off prompt alignment and diversity.<n>We trace this behavior to a previously overlooked source by decomposing total bias into two components: a model bias and a guidance bias.<n>We propose StayFair, which leverages this condition to design fair guidance algorithms in both regimes.
• Score: 12.234658039504728
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Diffusion models steer conditional generation with a tunable guidance scale to trade off prompt alignment and diversity. However, existing debiasing techniques are optimized for a single scale, degrading fairness when users adjust this parameter. We trace this behavior to a previously overlooked source by decomposing total bias into two components: a model bias and a guidance bias. While prior work primarily targets the former, we show that the guidance bias grows monotonically with the guidance scale, eventually dominating the high-guidance regimes users prefer. To address this, we extend Strong Demographic Parity to guidance and derive a condition under which the target distribution retains its group ratio across guidance scales. We propose StayFair, which leverages this condition to design fair guidance algorithms in both regimes. For classifier guidance, it equalizes the classifier's output distributions across groups; for classifier-free guidance, it shifts the null embedding by a prompt-dependent offset. Because StayFair modifies only the guidance step, it is orthogonal to model debiasing and can be layered onto existing fair diffusion models to extend their fairness across guidance scales. Across class-conditional and text-to-image generation, StayFair decouples fairness from the guidance scale without sacrificing image quality.

Related papers

• Fairness-Aware Mixture-of-Experts via Subgroup Reweighting and Gate Regularization [2.6245086561385276]
  We propose an end-to-end Mixture-of-Experts (MoE) framework that corrects subgroup imbalance.<n>We apply subgroup reweighting to correct data imbalance, and introduce gate entropy regularization to prevent routing from collapsing onto subgroup attributes.<n> Experimental results demonstrate that the proposed approach improves fairness while maintaining competitive predictive performance.
  arXiv  Detail & Related papers  (2026-08-24T05:34:25Z)
• Dual-Branch Cross-Projection Debiasing through Diffusion-based Disentanglement [66.67463557274358]
  Foundation models trained on biased datasets often rely on spurious correlations between target labels and non-causal attributes.<n>We introduce Confidence-guided Concept Mining (CBCM) to identify reliable spurious attributes without attribute annotations.<n>We propose Dual-branch Cross-projection Debiasing (DCD), a prompt-tuning framework that separates target and spurious representations into two branches.
  arXiv  Detail & Related papers  (2026-06-23T05:28:47Z)
• FairGen: Controlling Sensitive Attributes for Fair Generations in Diffusion Models via Adaptive Latent Guidance [22.603461883941495]
  Text-to-image diffusion models often exhibit biases toward specific demographic groups.<n>In this paper, we tackle the challenge of mitigating generation bias towards any target attribute value.<n>We propose FairGen, an adaptive latent guidance mechanism which controls the generation distribution during inference.
  arXiv  Detail & Related papers  (2025-02-25T23:47:22Z)
• FairGen: Enhancing Fairness in Text-to-Image Diffusion Models via Self-Discovering Latent Directions [18.3214925434309]
  Diffusion Models (DMs) exhibit remarkable performance across various image generative tasks.<n>These biases could perpetuate a distorted worldview and hinder opportunities for minority groups.<n>We propose FairGen, a plug-and-play method that learns attribute latent directions in a self-discovering manner.
  arXiv  Detail & Related papers  (2024-12-25T07:30:20Z)
• Finetuning Text-to-Image Diffusion Models for Fairness [43.80733100304361]
  We frame fairness as a distributional alignment problem. Empirically, our method markedly reduces gender, racial, and their intersectional biases for occupational prompts. Our method supports diverse perspectives of fairness beyond absolute equality.
  arXiv  Detail & Related papers  (2023-11-11T05:40:54Z)
• Fair-CDA: Continuous and Directional Augmentation for Group Fairness [48.84385689186208]
  We propose a fine-grained data augmentation strategy for imposing fairness constraints. We show that group fairness can be achieved by regularizing the models on transition paths of sensitive features between groups. Our proposed method does not assume any data generative model and ensures good generalization for both accuracy and fairness.
  arXiv  Detail & Related papers  (2023-04-01T11:23:00Z)
• DualFair: Fair Representation Learning at Both Group and Individual Levels via Contrastive Self-supervision [73.80009454050858]
  This work presents a self-supervised model, called DualFair, that can debias sensitive attributes like gender and race from learned representations. Our model jointly optimize for two fairness criteria - group fairness and counterfactual fairness.
  arXiv  Detail & Related papers  (2023-03-15T07:13:54Z)
• On Comparing Fair Classifiers under Data Bias [42.43344286660331]
  We study the effect of varying data biases on the accuracy and fairness of fair classifiers. Our experiments show how to integrate a measure of data bias risk in the existing fairness dashboards for real-world deployments.
  arXiv  Detail & Related papers  (2023-02-12T13:04:46Z)
• Domain Adaptation meets Individual Fairness. And they get along [48.95808607591299]
  We show that algorithmic fairness interventions can help machine learning models overcome distribution shifts. In particular, we show that enforcing suitable notions of individual fairness (IF) can improve the out-of-distribution accuracy of ML models.
  arXiv  Detail & Related papers  (2022-05-01T16:19:55Z)
• Self-Conditioned Generative Adversarial Networks for Image Editing [61.50205580051405]
  Generative Adversarial Networks (GANs) are susceptible to bias, learned from either the unbalanced data, or through mode collapse. We argue that this bias is responsible not only for fairness concerns, but that it plays a key role in the collapse of latent-traversal editing methods when deviating away from the distribution's core.
  arXiv  Detail & Related papers  (2022-02-08T18:08:24Z)
• Relieving Long-tailed Instance Segmentation via Pairwise Class Balance [85.53585498649252]
  Long-tailed instance segmentation is a challenging task due to the extreme imbalance of training samples among classes. It causes severe biases of the head classes (with majority samples) against the tailed ones. We propose a novel Pairwise Class Balance (PCB) method, built upon a confusion matrix which is updated during training to accumulate the ongoing prediction preferences.
  arXiv  Detail & Related papers  (2022-01-08T07:48:36Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.