Cognitive Biases in Large Language Models: A Survey and Mitigation Experiments

TL;DR

This paper surveys cognitive biases in large language models, evaluates mitigation techniques inspired by human studies, and finds that the AwaRe method effectively reduces biases in LLM outputs, improving rationality.

Contribution

It provides the first comprehensive survey of LLM cognitive biases and experimentally evaluates two mitigation methods, highlighting AwaRe's effectiveness.

Findings

AwaRe reduces bias influence in GPT-3.5 and GPT-4

SoPro has minimal impact on bias mitigation

Survey identifies key biases and mitigation challenges in LLMs

Abstract

Large Language Models (LLMs) are trained on large corpora written by humans and demonstrate high performance on various tasks. However, as humans are susceptible to cognitive biases, which can result in irrational judgments, LLMs can also be influenced by these biases, leading to irrational decision-making. For example, changing the order of options in multiple-choice questions affects the performance of LLMs due to order bias. In our research, we first conducted an extensive survey of existing studies examining LLMs' cognitive biases and their mitigation. The mitigation techniques in LLMs have the disadvantage that they are limited in the type of biases they can apply or require lengthy inputs or outputs. We then examined the effectiveness of two mitigation methods for humans, SoPro and AwaRe, when applied to LLMs, inspired by studies in crowdsourcing. To test the effectiveness of these methods, we conducted experiments on GPT-3.5 and GPT-4 to evaluate the influence of six biases on the outputs before and after applying these methods. The results demonstrate that while SoPro has little effect, AwaRe enables LLMs to mitigate the effect of these biases and make more rational responses.