FairPair: A Robust Evaluation of Biases in Language Models through Paired Perturbations

arXiv:2404.06619v1 Announce Type: cross
Abstract: The accurate evaluation of differential treatment in language models to specific groups is critical to ensuring a positive and safe user experience. An ideal evaluation should have the properties of being robust, extendable to new groups or attributes, and being able to capture biases that appear in typical usage (rather than just extreme, rare cases). Relatedly, bias evaluation should surface not only egregious biases but also ones that are subtle and commonplace, such as a …

abstract arxiv biases cs.cl cs.cy cs.lg differential evaluation experience language language models positive robust safe through treatment type