# ReST-RL: Reinforcing LLM Reasoning through Unified Self-Training and Value-Guided Search

arXiv: https://arxiv.org/abs/2508.19576 · HTML: https://mlsanity.com/paper/2508.19576/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Placed in Reinforcement learning for reasoning, one of 17 directions in LLM reasoning.
- Builds on 4 of its 38 references that are in the map.
- First posted on arXiv 27 Aug 2025.

## Placed in

- LLM reasoning / [Reinforcement learning for reasoning](https://mlsanity.com/lr/reinforcement-learning-for-reasoning/)
- Synthetic data and self-improvement for LLMs / [Self-play RL for reasoning](https://mlsanity.com/synth/self-play-rl-for-reasoning/)

## Builds on (most cited first, 4 of 4)

- [Qwen2.5-Coder Technical Report](https://mlsanity.com/paper/2409.12186/) (2024)
- [Beyond Human Data: Scaling Self-Training for Problem-Solving with Language Models](https://mlsanity.com/paper/2312.06585/) (2023)
- [Skywork-Reward: Bag of Tricks for Reward Modeling in LLMs](https://mlsanity.com/paper/2410.18451/) (2024)
- [OpenCodeInterpreter: Integrating Code Generation with Execution and Refinement](https://mlsanity.com/paper/2402.14658/) (2024)

## Built on by (most cited first, 9 of 9)

- [CostBench: Evaluating Multi-Turn Cost-Optimal Planning and Adaptation in Dynamic Environments for LLM Tool-Use Agents](https://mlsanity.com/paper/2511.02734/) (2025)
- [Spark: Strategic Policy-Aware Exploration via Dynamic Branching for Long-Horizon Agentic Learning](https://mlsanity.com/paper/2601.20209/) (2026)
- [Advancing General-Purpose Reasoning Models with Modular Gradient Surgery](https://mlsanity.com/paper/2602.02301/) (2026)
- [Beyond Unimodal Shortcuts: MLLMs as Cross-Modal Reasoners for Grounded Named Entity Recognition](https://mlsanity.com/paper/2602.04486/) (2026)
- [CLIPO: Contrastive Learning in Policy Optimization Generalizes RLVR](https://mlsanity.com/paper/2603.10101/) (2026)
- [How Many Code and Test Cases Are Enough? Evaluating Test Cases Generation from a Binary-Matrix Perspective](https://mlsanity.com/paper/2510.08720/) (2025)
- [MARTI-MARS2: Scaling Multi-Agent Self-Search via Reinforcement Learning for Code Generation](https://mlsanity.com/paper/2602.07848/) (2026)
- [MARS2: Scaling Multi-Agent Tree Search via Reinforcement Learning for Code Generation](https://mlsanity.com/paper/2604.14564/) (2026)
- [Continual Learning in Transition](https://mlsanity.com/paper/2608.06216/) (2026)

## Abstract (arXiv)

With respect to improving the reasoning accuracy of LLMs, the representative reinforcement learning (RL) method - Group Relative Policy Optimization (GRPO) - has achieved critical success, yet it still suffers from the issue of insignificant reward signals. This paper introduces ReST-RL, a unified Reinforced Self-Training (ReST) policy-value framework that reconnects policy optimization and value-guided search to improve LLM reasoning ability. Firstly, ReST-GRPO adopts an optimized ReST-style algorithm to reshape the policy-induced trajectory distribution by increasing the reward variance of GRPO sampling and exposing the policy to more informative partial states, thereby improving training efficiency and effectiveness. Then, we further introduce a decoding optimization method, VM-MCTS, which trains a Value Model (VM) from self-collected Monte-Carlo Tree Search (MCTS) targets and deploys it through an adapted MCTS algorithm to provide precise process signals and verification scores, further enhancing reasoning accuracy. These two stages are internally dependent - ReST-GRPO yields higher-quality trajectories for value learning with VM-MCTS, which in turn enables more effective inference-time search. We validate our framework on multiple coding benchmarks (e.g., APPS, BigCodeBench, and HumanEval), where it significantly outperforms other reinforcement training baselines (naive GRPO, DAPO, and ReST-DPO), as well as decoding and verification baselines (e.g., PRM-BoN and ORM-MCTS), indicating its power to strengthen LLM reasoning capability. Moreover, we further evaluate ReST-RL on out-of-domain math and science reasoning tasks, where it achieves improved performance without target-domain tuning and favorable end-to-end efficiency trade-offs, providing preliminary transfer evidence beyond our primary coding domain.