Alleviating Choice Supportive Bias in LLM with Reasoning Dependency Generation

TL;DR

This paper introduces a novel framework called Reasoning Dependency Generation (RDG) to mitigate choice-supportive bias in large language models by generating unbiased reasoning data for fine-tuning, significantly improving bias reduction while maintaining standard performance.

Contribution

The paper presents the first method to address cognitive biases in LLMs using RDG, which constructs balanced reasoning QA pairs to reduce choice-supportive bias through fine-tuning.

Findings

81.5% improvement in memory-based bias experiments

94.3% improvement in evaluation-based bias experiments

Maintains performance on standard benchmarks

Abstract

Recent studies have demonstrated that some Large Language Models exhibit choice-supportive bias (CSB) when performing evaluations, systematically favoring their chosen options and potentially compromising the objectivity of AI-assisted decision making. While existing debiasing approaches primarily target demographic and social biases, methods for addressing cognitive biases in LLMs remain largely unexplored. In this work, we present the first solution to address CSB through Reasoning Dependency Generation (RDG), a novel framework for generating unbiased reasoning data to mitigate choice-supportive bias through fine-tuning. RDG automatically constructs balanced reasoning QA pairs, explicitly (un)modeling the dependencies between choices, evidences, and justifications. Our approach is able to generate a large-scale dataset of QA pairs across domains, incorporating Contextual Dependency Data and Dependency Decouple Data. Experiments show that LLMs fine-tuned on RDG-generated data demonstrate a 81.5% improvement in memory-based experiments and 94.3% improvement in the evaluation-based experiment, while maintaining similar performance on standard BBQ benchmarks. This work pioneers an approach for addressing cognitive biases in LLMs and contributes to the development of more reliable AI-assisted decision support systems.