AI Chat AI Image Generator AI Video Text to Speech

Selection Bias Induced Spurious Correlations in Large Language Models

07/18/2022

∙

by Emily McMilin, et al.

∙

∙

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.

page 1

page 2

page 3

page 4

research

∙ 08/22/2022

Selection Collider Bias in Large Language Models

In this paper we motivate the causal mechanisms behind sample selection ...

0 Emily McMilin, et al. ∙

research

∙ 09/30/2022

Exploiting Selection Bias on Underspecified Tasks in Large Language Models

In this paper we motivate the causal mechanisms behind sample selection ...

0 Emily McMilin, et al. ∙

research

∙ 11/15/2021

Assessing gender bias in medical and scientific masked language models with StereoSet

NLP systems use language models such as Masked Language Models (MLMs) th...

0 Robert Robinson, et al. ∙

research

∙ 03/16/2023

MultiModal Bias: Introducing a Framework for Stereotypical Bias Assessment beyond Gender and Race in Vision Language Models

Recent breakthroughs in self supervised training have led to a new class...

0 Sepehr Janghorbani, et al. ∙

research

∙ 03/17/2023

She Elicits Requirements and He Tests: Software Engineering Gender Bias in Large Language Models

Implicit gender bias in software development is a well-documented issue,...

0 Christoph Treude, et al. ∙

research

∙ 11/01/2019

On the Unintended Social Bias of Training Language Generation Models with Data from Local Media

There are concerns that neural language models may preserve some of the ...

0 Omar U. Florez, et al. ∙

research

∙ 02/24/2023

In-Depth Look at Word Filling Societal Bias Measures

Many measures of societal bias in language models have been proposed in ...

0 Matúš Pikuliak, et al. ∙