Agents & Tool Use · 2025
SWE-RL: Advancing LLM Reasoning via Reinforcement Learning on Open Software Evolution
Uses rule-based rewards derived from GitHub commit histories to train issue-fixing coding agents by RL on real software evolution, without any proprietary model in the pipeline.
Editorial record
Plain-language summary
Uses rule-based rewards derived from GitHub commit histories to train issue-fixing coding agents by RL on real software evolution, without any proprietary model in the pipeline.
Knowledge graph
Relationships
Descendants
Conceptual ancestorEvidence: Strongly supported
ReAct: Synergizing Reasoning and Acting in LMs
Builds on the agents lineage in the archive
freshness sweep 2026
Source record
Provenance
- Record ID
- P-603
- Record created
- 2026-07-13
- Last reviewed
- 2026-07-14
- Record version
- 2
- https://arxiv.org/abs/2502.18449
- arXiv:2502.18449
Citation caveat: Citation metadata is approximate and marked unverified in the source dataset.