Selection Bias Induced Spurious Correlations in Large Language Models

Code

• github.com/2dot71mily/spurious_correlations_icml_2022
  OfficialIn papernone★ 1

Abstract

In this work we show how large language models (LLMs) can learn statistical dependencies between otherwise unconditionally independent variables due to dataset selection bias. To demonstrate the effect, we developed a masked gender task that can be applied to BERT-family models to reveal spurious correlations between predicted gender pronouns and a variety of seemingly gender-neutral variables like date and location, on pre-trained (unmodified) BERT and RoBERTa large models. Finally, we provide an online demo, inviting readers to experiment further.

Tasks

Reproductions