Deep neural networks often learn unintended bias during training, which might have harmful
effects when deployed in realworld settings. This work surveys 214 papers related to …

Bias amplification is a phenomenon in which models increase imbalances present in the
training data. In this paper, we study bias amplification in the text-to-image domain using …

Robust, faithful, and harm-free pronoun use for individuals is an important goal for language
model development as their use increases, but prior work tends to study only one or two of …

Large language models (LLMs) are susceptible to a variety of risks, from non-faithful output
to biased and toxic generations. Due to several limiting factors surrounding LLMs (training …

Rapid advancements of large language models (LLMs) have enabled the processing,
understanding, and generation of human-like text, with increasing integration into systems …

Recent studies have shown that Text-to-Image (T2I) model generations can reflect social
stereotypes present in the real world. However, existing approaches for evaluating …

Current datasets for unwanted social bias auditing are limited to studying protected
demographic features such as race and gender. In this work, we introduce a comprehensive …

This paper introduces the concept of actionability in the context of bias measures in natural
language processing (NLP). We define actionability as the degree to which a …

Gender bias research has been pivotal in revealing undesirable behaviors in large
language models, exposing serious gender stereotypes associated with occupations, and …

Sociodemographic bias in language models (LMs) has the potential for harm when
deployed in real-world settings. This paper presents a comprehensive survey of the past …