用推理依赖生成法减轻大模型的选择性偏见，提升决策客观性。
Alleviating Choice Supportive Bias in LLM with Reasoning Dependency Generation
- 构建推理依赖生成框架，自动构造平衡的问答对
- 在记忆和评估实验中分别提升81.5%与94.3%的准确率
- 适合需要可靠决策支持的AI应用开发者
近期研究表明，部分大型语言模型在评估任务中存在选择性偏见（CSB），会系统性地偏好自身选择的选项，可能损害人机协同决策的客观性。现有去偏方法多针对人口与社会偏见，而针对认知偏见的解决方案仍不充分。本文提出首个通过推理依赖生成（RDG）缓解CSB的方法，该框架可自动生成无偏见的推理数据以用于微调。RDG能自动构建跨领域的平衡推理问答对，显式建模或解耦选择、证据与理由之间的依赖关系。所生成的数据集包含上下文依赖数据与依赖解耦数据。实验表明，基于RDG数据微调的模型在记忆型实验中性能提升81.5%，在评估型实验中提升94.3%，同时在标准BBQ基准上保持相近表现。本工作首次探索了大模型认知偏见的缓解路径，推动更可靠的AI辅助决策系统发展。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。