S$^2$R: Teaching LLMs to Self-verify and Self-correct via Reinforcement Learning

ACL ARR 2025 February Submission7230 Authors

Abstract: Recent studies have demonstrated the effectiveness of LLM test-time scaling. However, existing approaches to incentivize LLMs' deep thinking abilities generally require large-scale data or significant training efforts. Meanwhile, it remains unclear how to improve the thinking abilities of less powerful base models. In this work, we introduce S$^2$R, an efficient framework that enhances LLM reasoning by teaching models to self-verify and self-correct during inference. Specifically, we first initialize LLMs with iterative self-verification and self-correction behaviors through supervised fine-tuning on carefully curated data. The self-verification and self-correction skills are then further strengthened by outcome-level and process-level reinforcement learning with minimized resource requirements. Our results demonstrate that, with only 3.1k behavior initialization samples, Qwen2.5-math-7B achieves an accuracy improvement from 51.0\% to 81.6\%, outperforming models trained on an equivalent amount of long-CoT distilled data. We also discuss the effect of different RL strategies on enhancing LLMs' deep reasoning. Extensive experiments and analysis based on three base models across both in-domain and out-of-domain benchmarks validate the effectiveness of S$^2$R.

Paper Type: Long

Research Area: NLP Applications

Research Area Keywords: Large Language Model, LLM Reasoning, Reinforcement Learning, Test-time Scaling

Contribution Types: Model analysis & interpretability, NLP engineering experiment, Approaches to low-resource settings, Approaches low compute settings-efficiency, Publicly available software and/or pre-trained models

Languages Studied: English

Submission Number: 7230