【论文速递】StepBack: Step-Level Error-Localized Resampling for Efficient Test-Time Reasonin

英文标题：StepBack: Step-Level Error-Localized Resampling for Efficient Test-Time Reasoning

arXiv编号：2611.08403
作者：Yang Ouyang, Jung-Eun Kim
来源：arXiv
原文链接：https://arxiv.org/abs/2611.08403

中文摘要：
Self-consistency (SC) is a widely used decoding strategy for improving chain-of-thought reasoning, but it requires sampling many complete reasoning trajectories with a preset budget. Its efficient variants reduce the cost mainly by stopping full-trajectory sampling early or distributing the number of samples across questions. In this paper, we propose StepBack, a step-level error-localized resampling method for efficient test-time reasoning. Given a small set of draft trajectories, StepBack localizes likely first error positions using lightweight signals and then resamples the detected error step and its suffix, followed by majority answer voting. Empirically, StepBack matches the accuracy of SC on average across seven mathematical reasoning and multi-hop QA benchmarks, while reducing token usage by up to 41.26%. In addition, by nature, StepBack can be combined with adaptive inference methods such as Adaptive Consistency, further improving their inference-time efficiency.

本文由华人空间 AI 论文助手自动抓取翻译，仅供学术交流。