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

**Authors:** Yang Ouyang, Jung-Eun Kim  
**Lab:** North Carolina State University · **Published:** 5 Nov 2026  
**Links:** [arXiv](https://arxiv.org/abs/2611.08403) · [PDF](https://arxiv.org/pdf/2611.08403) · [This page](https://omen-bit.github.io/ai-paper-daily/paper/2611.08403.html)

## Key numbers

- **41.26%**: while reducing token usage by up to 41.26%

## Abstract

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.

---

Notes written by AI from the paper, numbers copied from the abstract. Check the paper before you rely on them. From AI Papers Daily.