# Training Long-Context, Multi-Turn Software Engineering Agents with Reinforcement Learning

arXiv: https://arxiv.org/abs/2508.03501 · HTML: https://mlsanity.com/paper/2508.03501/

## Key facts

- Cited by 36 papers in the map (data through 14 Oct 2026).
- Placed in Reinforcement learning for reasoning, one of 17 directions in LLM reasoning.
- Builds on 19 of its 33 references that are in the map.
- First posted on arXiv 5 Aug 2025.

## Placed in

- LLM reasoning / [Reinforcement learning for reasoning](https://mlsanity.com/lr/reinforcement-learning-for-reasoning/)
- LLM agents and tool use / [Reinforcement learning for LLM agents](https://mlsanity.com/agents/reinforcement-learning-for-llm-agents/)

## Builds on (most cited first, 19 of 19)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [SWE-bench: Can Language Models Resolve Real-World GitHub Issues?](https://mlsanity.com/paper/2310.06770/) (2023)
- [DAPO: An Open-Source LLM Reinforcement Learning System at Scale](https://mlsanity.com/paper/2503.14476/) (2025)
- [SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering](https://mlsanity.com/paper/2405.15793/) (2024)
- [OpenHands: An Open Platform for AI Software Developers as Generalist Agents](https://mlsanity.com/paper/2407.16741/) (2024)
- [YaRN: Efficient Context Window Extension of Large Language Models](https://mlsanity.com/paper/2309.00071/) (2023)
- [Agentless: Demystifying LLM-based Software Engineering Agents](https://mlsanity.com/paper/2407.01489/) (2024)
- [Training Software Engineering Agents and Verifiers with SWE-Gym](https://mlsanity.com/paper/2412.21139/) (2024)
- [R2E-Gym: Procedural Environments and Hybrid Verifiers for Scaling Open-Weights SWE Agents](https://mlsanity.com/paper/2504.07164/) (2025)
- [SWE-rebench: An Automated Pipeline for Task Collection and Decontaminated Evaluation of Software Engineering Agents](https://mlsanity.com/paper/2505.20411/) (2025)
- [SWE-Search: Enhancing Software Agents with Monte Carlo Tree Search and Iterative Refinement](https://mlsanity.com/paper/2410.20285/) (2024)
- [Curriculum Reinforcement Learning from Easy to Hard Tasks Improves LLM Reasoning](https://mlsanity.com/paper/2506.06632/) (2025)
- [SWE-Fixer: Training Open-Source LLMs for Effective and Efficient GitHub Issue Resolution](https://mlsanity.com/paper/2501.05040/) (2025)
- [StepCoder: Improve Code Generation with Reinforcement Learning from Compiler Feedback](https://mlsanity.com/paper/2402.01391/) (2024)
- [Lingma SWE-GPT: An Open Development-Process-Centric Language Model for Automated Software Improvement](https://mlsanity.com/paper/2411.00622/) (2024)
- [Guided Search Strategies in Non-Serializable Environments with Applications to Software Engineering Agents](https://mlsanity.com/paper/2505.13652/) (2025)

## Built on by (most cited first, 30 of 36)

- [The Landscape of Agentic Reinforcement Learning for LLMs: A Survey](https://mlsanity.com/paper/2509.02547/) (2025)
- [UserRL: Training Interactive User-Centric Agent via Reinforcement Learning](https://mlsanity.com/paper/2509.19736/) (2025)
- [SWE-World: Building Software Engineering Agents in Docker-Free Environments](https://mlsanity.com/paper/2602.03419/) (2026)
- [CUA-Gym: Scaling Verifiable Training Environments and Tasks for Computer-Use Agents](https://mlsanity.com/paper/2605.25624/) (2026)
- [SWE-Master: Unleashing the Potential of Software Engineering Agents via Post-Training](https://mlsanity.com/paper/2602.03411/) (2026)
- [Immersion in the GitHub Universe: Scaling Coding Agents to Mastery](https://mlsanity.com/paper/2602.09892/) (2026)
- [Composer 2 Technical Report](https://mlsanity.com/paper/2603.24477/) (2026)
- [Self-Improvements in Modern Agentic Systems: A Survey](https://mlsanity.com/paper/2607.13104/) (2026)
- [The Best of N Worlds: Aligning Reinforcement Learning with Best-of-N Sampling via max@k Optimisation](https://mlsanity.com/paper/2510.23393/) (2025)
- [DeltaBox: Scaling Stateful AI Agents with Millisecond-Level Sandbox Checkpoint/Rollback](https://mlsanity.com/paper/2605.22781/) (2026)
- [Supervised Reinforcement Learning: From Expert Trajectories to Step-wise Reasoning](https://mlsanity.com/paper/2510.25992/) (2025)
- [VulnLLM-R: Specialized Reasoning LLM with Agent Scaffold for Vulnerability Detection](https://mlsanity.com/paper/2512.07533/) (2025)
- [Dockerless: Environment-Free Program Verifier for Coding Agents](https://mlsanity.com/paper/2606.28436/) (2026)
- [Agentic Software Issue Resolution with Large Language Models: A Survey](https://mlsanity.com/paper/2512.22256/) (2025)
- [From SWE-ZERO to SWE-HERO: Execution-free to Execution-based Fine-tuning for Software Engineering Agents](https://mlsanity.com/paper/2604.01496/) (2026)
- [Fast LLM Post-training via Decoupled and Fastest-of-N Speculation](https://mlsanity.com/paper/2511.16193/) (2025)
- [AceGRPO: Adaptive Curriculum Enhanced Group Relative Policy Optimization for Autonomous Machine Learning Engineering](https://mlsanity.com/paper/2602.07906/) (2026)
- [Mach-Mind-4-Flash Technical Report](https://mlsanity.com/paper/2607.09375/) (2026)
- [LEGO-RL: Harness-Native Reinforcement Learning for Coding Agents](https://mlsanity.com/paper/2608.17393/) (2026)
- [FrogNano: Training a 4B Coding Agent via Online Task Synthesis](https://mlsanity.com/paper/2609.07925/) (2026)
- [ExecCritic: Learn to Test, Test to Improve for Coding Agents](https://mlsanity.com/paper/2609.09133/) (2026)
- [RepoLaunch: Automating Build and Management of Code Repositories across Languages and Platforms](https://mlsanity.com/paper/2603.05026/) (2026)
- [Function-Aware Fill-in-the-Middle as Mid-Training for Coding Agent Foundation Models](https://mlsanity.com/paper/2607.12463/) (2026)
- [ORACLE-SWE: Quantifying the Contribution of Oracle Information Signals on SWE Agents](https://mlsanity.com/paper/2604.07789/) (2026)
- [Rethinking Agentic Reinforcement Learning In Large Language Models](https://mlsanity.com/paper/2604.27859/) (2026)
- [Terminal Agents: A Survey of AI Agents in Command-Line Environments](https://mlsanity.com/paper/2608.20485/) (2026)
- [PIPer: On-Device Environment Setup via Online Reinforcement Learning](https://mlsanity.com/paper/2509.25455/) (2025)
- [EARL: Efficient Agentic Reinforcement Learning Systems for Large Language Models](https://mlsanity.com/paper/2510.05943/) (2025)
- [GeoMind: An Agentic Workflow for Lithology Classification with Reasoned Tool Invocation](https://mlsanity.com/paper/2604.21501/) (2026)
- [Signal Reshaping for GRPO in Weak-Feedback Agentic Code Repair](https://mlsanity.com/paper/2605.07276/) (2026)

## Abstract (arXiv)

Research on applications of reinforcement learning (RL) to large language models has mostly been focused on single-turn problems, such as mathematical reasoning or single-shot code generation. While these problems can be viewed as token-level multi-turn Markov decision processes (MDPs), this view corresponds to a degenerate case of multi-turn interaction where the environment provides no feedback. This contrasts with many real-world domains, such as software engineering (SWE), which require rich multi-turn interactions with a stateful environment that responds to each action with a non-trivial observation. To bridge this gap, we demonstrate the successful application of RL to this general regime. Our methodology begins with rejection fine-tuning (RFT) using execution feedback to train a policy to follow instructions and formatting effectively, followed by a synchronous RL pipeline using DAPO for iterative improvement. Applying this pipeline to Qwen2.5-72B-Instruct, we increase its Pass@1 on the SWE-bench Verified benchmark from 11% to 39%, substantially improving upon the 20% RFT baseline. On the May and June splits of SWE-rebench, the resulting agent achieves Pass@1 of 35% and 31% respectively, competitive with even larger models such as DeepSeek-V3-0324 or Qwen3-235B-A22B, demonstrating that our methodology offers a practical approach for training capable agents for multi-turn interactive tasks using open-weight models.