DebFilter: Eradicating Biases Stashed in Value

• URL: http://arxiv.org/abs/2605.28167v1
• Date: Wed, 27 May 2026 08:49:41 GMT
• Title: DebFilter: Eradicating Biases Stashed in Value
• Abstract summary: We propose DebFilter to mitigate social and semantic biases in text-to-image models.<n>We apply a bias-correction strategy that adjusts the value components within cross-attention.<n>Our results demonstrate that this method effectively reconfigures social biases in generated images.
• Score: 4.060731229044571
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Text-to-image diffusion models, which are theoretically equivalent to score-based generative models, generate images through a multi-step denoising process guided by text embeddings extracted from pretrained vision-language models such as CLIP. However, these text embeddings inherently encode social and semantic biases -- such as those related to gender and age -- that are subsequently propagated and amplified through the guidance mechanism, along with the model's training on large-scale datasets that are imbalanced with respect to these bias-related concepts, often leading to skewed outputs in text-to-image generation. We propose DebFilter, a lightweight and training-free framework for mitigating such biases in text-to-image diffusion models. Observing that the model's error prediction at each denoising step is primarily influenced by cross-attention dynamics, we introduce a bias-correction strategy that adjusts the value components within cross-attention. Specifically, we apply a fixed offset to the slice of guidance embedding, effectively steering the semantic direction of cross-attention values toward unbiased representations. This adjustment reconfigures the score landscape to produce balanced outputs while maintaining alignment with the intended text semantics. Unlike prior approaches that rely on fine-tuning or retraining, DebFilter operates entirely at inference time, requiring no additional data or model updates. Our results demonstrate that this method effectively mitigates social biases in generated images, offering an efficient and scalable pathway toward fairer and more inclusive text-to-image generation.

Related papers

• In-Loop Model Adaptation with Coupled Latent-Noise Guidance for High-Fidelity Subject-Driven Text-to-Image Generation [43.97733750301509]
  We propose a new approach for subject-driven text-to-image generation called textitIn-Loop Model Adaptation (IMA)<n>IMA adapts the core diffusion model at each generation step during the actual process of image generation without being trained on the reference image before the generation process.<n>Our experiments demonstrate that our proposed IMA method significantly improves the performance of subject-driven text-to-image generation.
  arXiv  Detail & Related papers  (2026-08-10T08:07:28Z)
• FairImagen: Post-Processing for Bias Mitigation in Text-to-Image Models [10.857020427374506]
  We introduce FairImagen, a post-hoc debiasing framework that operates on prompt embeddings to mitigate societal biases.<n>Our framework outperforms existing post-hoc methods and offers a simple, scalable, and model-agnostic solution for equitable text-to-image generation.
  arXiv  Detail & Related papers  (2025-10-24T11:47:15Z)
• EditInfinity: Image Editing with Binary-Quantized Generative Models [64.05135380710749]
  We investigate the parameter-efficient adaptation of binary-quantized generative models for image editing.<n>Specifically, we propose EditInfinity, which adapts emphInfinity, a binary-quantized generative model, for image editing.<n>We propose an efficient yet effective image inversion mechanism that integrates text prompting rectification and image style preservation.
  arXiv  Detail & Related papers  (2025-10-23T05:06:24Z)
• Debiasing Vison-Language Models with Text-Only Training [15.069736314663352]
  We propose a Text-Only Debiasing framework called TOD, leveraging a text-as-image training paradigm to mitigate visual biases. To address the limitations, we propose a Text-Only Debiasing framework called TOD, leveraging a text-as-image training paradigm to mitigate visual biases.
  arXiv  Detail & Related papers  (2024-10-12T04:34:46Z)
• Removing Distributional Discrepancies in Captions Improves Image-Text Alignment [76.31530836622694]
  We introduce a model designed to improve the prediction of image-text alignment. Our approach focuses on generating high-quality training datasets for the alignment task. We also demonstrate the applicability of our model by ranking the images generated by text-to-image models based on text alignment.
  arXiv  Detail & Related papers  (2024-10-01T17:50:17Z)
• AITTI: Learning Adaptive Inclusive Token for Text-to-Image Generation [53.65701943405546]
  We learn adaptive inclusive tokens to shift the attribute distribution of the final generative outputs. Our method requires neither explicit attribute specification nor prior knowledge of the bias distribution. Our method achieves comparable performance to models that require specific attributes or editing directions for generation.
  arXiv  Detail & Related papers  (2024-06-18T17:22:23Z)
• Enhancing Semantic Fidelity in Text-to-Image Synthesis: Attention Regulation in Diffusion Models [23.786473791344395]
  Cross-attention layers in diffusion models tend to disproportionately focus on certain tokens during the generation process. We introduce attention regulation, an on-the-fly optimization approach at inference time to align attention maps with the input text prompt. Experiment results show that our method consistently outperforms other baselines.
  arXiv  Detail & Related papers  (2024-03-11T02:18:27Z)
• Fair Text-to-Image Diffusion via Fair Mapping [32.02815667307623]
  We propose a flexible, model-agnostic, and lightweight approach that modifies a pre-trained text-to-image diffusion model. By effectively addressing the issue of implicit language bias, our method produces more fair and diverse image outputs.
  arXiv  Detail & Related papers  (2023-11-29T15:02:01Z)
• Steered Diffusion: A Generalized Framework for Plug-and-Play Conditional Image Synthesis [62.07413805483241]
  Steered Diffusion is a framework for zero-shot conditional image generation using a diffusion model trained for unconditional generation. We present experiments using steered diffusion on several tasks including inpainting, colorization, text-guided semantic editing, and image super-resolution.
  arXiv  Detail & Related papers  (2023-09-30T02:03:22Z)
• Balancing the Picture: Debiasing Vision-Language Datasets with Synthetic Contrast Sets [52.77024349608834]
  Vision-language models can perpetuate and amplify societal biases learned during pre-training on uncurated image-text pairs from the internet. COCO Captions is the most commonly used dataset for evaluating bias between background context and the gender of people in-situ. We propose a novel dataset debiasing pipeline to augment the COCO dataset with synthetic, gender-balanced contrast sets.
  arXiv  Detail & Related papers  (2023-05-24T17:59:18Z)
• Debiasing Vision-Language Models via Biased Prompts [79.04467131711775]
  We propose a general approach for debiasing vision-language foundation models by projecting out biased directions in the text embedding. We show that debiasing only the text embedding with a calibrated projection matrix suffices to yield robust classifiers and fair generative models.
  arXiv  Detail & Related papers  (2023-01-31T20:09:33Z)

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.