Do Biased Models Have Biased Thoughts?

Swati Rajwal, Shivank Garg, Reem Abdel-Salam, Abdelrahman Zayed

Keywords: Bias in language models, Large Language Models, biased thoughts, Chain-of-Thought prompting

TL;DR: This paper explores whether biased language models have biased reasoning, finding that their thought processes are not strongly linked to biased outputs.

Abstract: The impressive performance of language models is undeniable. However, the presence of biases based on gender, race, socio-economic status, physical appearance, and sexual orientation makes the deployment of language models challenging. This paper studies the effect of chain-of-thought prompting, a recent approach that studies the steps followed by the model before it responds, on fairness. More specifically, we ask the following question: *Do biased models have biased thoughts*? To answer our question, we conduct experiments on $5$ popular large language models using fairness metrics to quantify $11$ different biases in the model's thoughts and output. Our results show that the bias in the thinking steps is not highly correlated with the output bias (less than $0.6$ correlation with a $p$-value smaller than $0.001$ in most cases). In other words, unlike human beings, the tested models with biased decisions do not always possess biased thoughts.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the COLM Code of Ethics on https://colmweb.org/CoE.html

Author Guide: I certify that this submission complies with the submission instructions as described on https://colmweb.org/AuthorGuide.html

Submission Number: 203