Selection Collider Bias in Large Language Models

Emily McMilin

Keywords: large language models, causal inference, selection bias

TL;DR: Using causal inference methods, we explain and demonstrate how sample selection bias causes spurious correlations during training, and how those spurious correlations can be used to classify prediction tasks as underspecified during inference.

Abstract: In this paper we motivate the causal mechanisms behind sample selection induced collider bias (selection collider bias) that can cause Large Language Mod- els (LLMs) to learn unconditional dependence between entities that are unconditionally independent in the real world. We show that selection collider bias can become amplified in underspecified learning tasks, and although difficult to overcome, we describe a method to exploit the resulting spurious correlations for determination of when a model may be uncertain about its prediction. We demonstrate an uncertainty metric that matches human uncertainty in tasks with gender pronoun underspecification on an extended version of the Winogender Schemas evaluation set, and we provide online demos where users can evaluate spurious correlations and apply our uncertainty metric to their own texts and models. Finally, we generalize our approach to address a wider range of prediction tasks.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Submission Guidelines: Yes

Please Choose The Closest Area That Your Submission Falls Into: Probabilistic Methods (eg, variational inference, causal inference, Gaussian processes)

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/arxiv:2208.10063/code)