FairRAG: Fair Human Generation via Fair Retrieval Augmentation

arXiv:2403.19964v1 Announce Type: cross
Abstract: Existing text-to-image generative models reflect or even amplify societal biases ingrained in their training data. This is especially concerning for human image generation where models are biased against certain demographic groups. Existing attempts to rectify this issue are hindered by the inherent limitations of the pre-trained models and fail to substantially improve demographic diversity. In this work, we introduce Fair Retrieval Augmented Generation (FairRAG), a novel framework that conditions pre-trained generative models on reference images …

abstract amplify arxiv augmentation biases cs.cv cs.cy cs.lg data fair generative generative models human image image generation issue limitations pre-trained models retrieval text text-to-image training training data type via