让大模型分步评估自身推理，提升复杂任务中的错误检测能力
Self-Evaluating LLMs for Multi-Step Tasks: Stepwise Confidence Estimation for Failure Detection
- 分步评估比整体评估更有效，逐环节判断可靠性
- 在多步任务中，错误检测的AUC-ROC最高提升15%相对值
- 适合需要高可靠性的复杂推理场景，如医疗、金融
大语言模型在高风险多步推理任务中的可靠性与故障检测至关重要。现有方法多聚焦单步输出的置信度估计，忽视多步推理的挑战。本文将自评估技术扩展至多步任务，测试整体评分与分步评分两种策略。基于两个多步基准数据集的实验表明，分步评估在错误检测上普遍优于整体评分，AUC-ROC最高提升15%相对值。结果证明，自评估大模型系统能在复杂推理中提供有意义的置信度估计，增强可信度，并为故障检测提供实用框架。
原文摘要 · Abstract (English)
Reliability and failure detection of large language models (LLMs) is critical for their deployment in high-stakes, multi-step reasoning tasks. Prior work explores confidence estimation for self-evaluating LLM-scorer systems, with confidence scorers estimating the likelihood of errors in LLM responses. However, most methods focus on single-step outputs and overlook the challenges of multi-step reasoning. In this work, we extend self-evaluation techniques to multi-step tasks, testing two intuitive approaches: holistic scoring and step-by-step scoring. Using two multi-step benchmark datasets, we show that stepwise evaluation generally outperforms holistic scoring in detecting potential errors, with up to 15% relative increase in AUC-ROC. Our findings demonstrate that self-evaluating LLM systems provide meaningful confidence estimates in complex reasoning, improving their trustworthiness and providing a practical framework for failure detection.
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。