---
title: "StepBack: Step-Level Error-Localized Resampling for Efficient Test-Time Reasoning"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2611.08403v1"
permalink: "https://feed.astralor.com/items/itm_8059dba96195410974f2caff"
published: "2026-11-05T18:12:49.000Z"
score: 57
channel: "tech"
tags: ["推理效率","RAG","StepBack","Step-Level Error-Localized Resampling","Efficient Test-Time Reasoning","Self-consistency","Its"]
---

# StepBack: Step-Level Error-Localized Resampling for Efficient Test-Time Reasoning

## At a glance

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 co…

## Key points

- 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.

## Full text

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.

---

Verify against the original before citing; curated by Astral Feed: https://feed.astralor.com/items/itm_8059dba96195410974f2caff