FairDeDup: Detecting and Mitigating Vision-Language Fairness Disparities in Semantic Dataset Deduplication

arXiv:2404.16123v1 Announce Type: new
Abstract: Recent dataset deduplication techniques have demonstrated that content-aware dataset pruning can dramatically reduce the cost of training Vision-Language Pretrained (VLP) models without significant performance losses compared to training on the original dataset. These results have been based on pruning commonly used image-caption datasets collected from the web -- datasets that are known to harbor harmful social biases that may then be codified in trained models. In this work, we evaluate how deduplication affects the prevalence …

abstract arxiv cost cs.ai cs.cl cs.cv dataset datasets deduplication fairness image language losses performance pruning reduce results semantic training type vision vision-language