# Step Back to Leap Forward: Self-Backtracking for Boosting Reasoning of Language Models

arXiv: https://arxiv.org/abs/2502.04404 · HTML: https://mlsanity.com/paper/2502.04404/

## Key facts

- Cited by 27 papers in the map (data through 14 Oct 2026).
- Placed in Test-time compute and efficient thinking, one of 17 directions in LLM reasoning.
- Builds on 17 of its 38 references that are in the map.
- First posted on arXiv 6 Feb 2025.

## Placed in

- LLM reasoning / [Test-time compute and efficient thinking](https://mlsanity.com/lr/test-time-compute-and-efficient-thinking/)

## Builds on (most cited first, 17 of 17)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters](https://mlsanity.com/paper/2408.03314/) (2024)
- [Self-Rewarding Language Models](https://mlsanity.com/paper/2401.10020/) (2024)
- [Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMs](https://mlsanity.com/paper/2412.21187/) (2024)
- [Generative Language Modeling for Automated Theorem Proving](https://mlsanity.com/paper/2009.03393/) (2020)
- [ReST-MCTS*: LLM Self-Training via Process Reward Guided Tree Search](https://mlsanity.com/paper/2406.03816/) (2024)
- [KTO: Model Alignment as Prospect Theoretic Optimization](https://mlsanity.com/paper/2402.01306/) (2024)
- [Making Large Language Models Better Reasoners with Step-Aware Verifier](https://mlsanity.com/paper/2206.02336/) (2022)
- [O1 Replication Journey: A Strategic Progress Report -- Part 1](https://mlsanity.com/paper/2410.18982/) (2024)
- [Learning From Mistakes Makes LLM Better Reasoner](https://mlsanity.com/paper/2310.20689/) (2023)
- [Can LLMs Learn from Previous Mistakes? Investigating LLMs' Errors to Boost for Reasoning](https://mlsanity.com/paper/2403.20046/) (2024)
- [Physics of Language Models: Part 2.2, How to Learn From Mistakes on Grade-School Math Problems](https://mlsanity.com/paper/2408.16293/) (2024)
- [Dualformer: Controllable Fast and Slow Thinking by Learning with Randomized Reasoning Traces](https://mlsanity.com/paper/2410.09918/) (2024)
- [Backtracking Improves Generation Safety](https://mlsanity.com/paper/2409.14586/) (2024)
- [Learning to Better Search with Language Models via Guided Reinforced Self-Training](https://mlsanity.com/paper/2410.02992/) (2024)

## Built on by (most cited first, 27 of 27)

- [Towards Reasoning Era: A Survey of Long Chain-of-Thought for Reasoning Large Language Models](https://mlsanity.com/paper/2503.09567/) (2025)
- [Large Language Models for Planning: A Comprehensive and Systematic Survey](https://mlsanity.com/paper/2505.19683/) (2025)
- [Where Did It All Go Wrong? A Hierarchical Look into Multi-Agent Error Attribution](https://mlsanity.com/paper/2510.04886/) (2025)
- [Reasoning-Finetuning Repurposes Latent Representations in Base Models](https://mlsanity.com/paper/2507.12638/) (2025)
- [Auditing Meta-Cognitive Hallucinations in Reasoning Large Language Models](https://mlsanity.com/paper/2505.13143/) (2025)
- [WebRollback: Enhancing Web Agents with Explicit Rollback Mechanisms](https://mlsanity.com/paper/2504.11788/) (2025)
- [NeuReasoner: Towards Explainable, Controllable, and Unified Reasoning via Mixture-of-Neurons](https://mlsanity.com/paper/2604.02972/) (2026)
- [How Much Backtracking is Enough? Exploring the Interplay of SFT and RL in Enhancing LLM Reasoning](https://mlsanity.com/paper/2505.24273/) (2025)
- [Modeling Student Learning with 3.8 Million Program Traces](https://mlsanity.com/paper/2510.05056/) (2025)
- [First Finish Search: Efficient Test-Time Scaling in Large Language Models](https://mlsanity.com/paper/2505.18149/) (2025)
- [Test-time Scaling Techniques in Theoretical Physics -- A Comparison of Methods on the TPBench Dataset](https://mlsanity.com/paper/2506.20729/) (2025)
- [Schoenfeld's Anatomy of Mathematical Reasoning by Language Models](https://mlsanity.com/paper/2512.19995/) (2025)
- [Re2: Unlocking LLM Reasoning via Reinforcement Learning with Re-solving](https://mlsanity.com/paper/2603.07197/) (2026)
- [Demystifying Scientific Problem-Solving in LLMs by Probing Knowledge and Reasoning](https://mlsanity.com/paper/2508.19202/) (2025)
- [How LLMs Detect and Correct Their Own Errors: The Role of Internal Confidence Signals](https://mlsanity.com/paper/2604.22271/) (2026)
- [Conformal Certification of Reasoning Trace Prefixes](https://mlsanity.com/paper/2605.30085/) (2026)
- [Provable Benefits of RLVR over SFT for Reasoning Models: Learning to Backtrack Efficiently](https://mlsanity.com/paper/2606.22938/) (2026)
- [STRIDE: Learnable Stepwise Language Feedback for LLM Reasoning](https://mlsanity.com/paper/2605.18851/) (2026)
- [Search, Fail, Recover: A Training Framework for Correction-Aware Reasoning](https://mlsanity.com/paper/2607.07492/) (2026)
- [CUSUM-Shaped Inference-Time Monitoring and Targeted Re-Decoding for Quantized Small Language Model Reasoning](https://mlsanity.com/paper/2607.20129/) (2026)
- [Adaptive Backtracking for Privacy Protection in Large Language Models](https://mlsanity.com/paper/2508.06087/) (2025)
- [Synthetic Error Injection Fails to Elicit Self-Correction In Language Models](https://mlsanity.com/paper/2512.02389/) (2025)
- [Caterpillar of Thoughts: The Optimal Test-Time Algorithm for Large Language Models](https://mlsanity.com/paper/2603.22784/) (2026)
- [Behavior Cue Reasoning: Monitorable Reasoning Improves Efficiency and Safety through Oversight](https://mlsanity.com/paper/2605.07021/) (2026)
- [The Periodic Table of LLM Reasoning: A Structured Survey of Reasoning Paradigms, Methods, and Failure Modes](https://mlsanity.com/paper/2606.11470/) (2026)
- [Calibrated e-CUSUM Decoding for Quantized Reasoning Models: Why Token Log-Probability Is the Wrong Observable for Decoding Monitors](https://mlsanity.com/paper/2607.11317/) (2026)
- [Eta Given Delta: Defining LLM Tool Efficiency With Marginal Tool Utility](https://mlsanity.com/paper/2607.14108/) (2026)

## Abstract (arXiv)

The integration of slow-thinking mechanisms into large language models (LLMs) offers a promising way toward achieving Level 2 AGI Reasoners, as exemplified by systems like OpenAI's o1. However, several significant challenges remain, including inefficient overthinking and an overreliance on auxiliary reward models. We point out that these limitations stem from LLMs' inability to internalize the search process, a key component of effective reasoning. A critical step toward addressing this issue is enabling LLMs to autonomously determine when and where to backtrack, a fundamental operation in traditional search algorithms. To this end, we propose a self-backtracking mechanism that equips LLMs with the ability to backtrack during both training and inference. This mechanism not only enhances reasoning ability but also efficiency by transforming slow-thinking processes into fast-thinking through self-improvement. Empirical evaluations demonstrate that our proposal significantly enhances the reasoning capabilities of LLMs, achieving a performance gain of over 40 percent compared to the optimal-path supervised fine-tuning method. We believe this study introduces a novel and promising pathway for developing more advanced and robust Reasoners.