SWE-Master: Unleashing the Potential of Software Engineering Agents via Post-Training
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917290042195968 |
|---|---|
| author | Song, Huatong Huang, Lisheng Sun, Shuang Jiang, Jinhao Le, Ran Cheng, Daixuan Chen, Guoxin Hu, Yiwen Chen, Zongchao Jia, Yiming Zhao, Wayne Xin Song, Yang Zhang, Tao Wen, Ji-Rong |
| author_facet | Song, Huatong Huang, Lisheng Sun, Shuang Jiang, Jinhao Le, Ran Cheng, Daixuan Chen, Guoxin Hu, Yiwen Chen, Zongchao Jia, Yiming Zhao, Wayne Xin Song, Yang Zhang, Tao Wen, Ji-Rong |
| contents | In this technical report, we present SWE-Master, an open-source and fully reproducible post-training framework for building effective software engineering agents. SWE-Master systematically explores the complete agent development pipeline, including teacher-trajectory synthesis and data curation, long-horizon SFT, RL with real execution feedback, and inference framework design. Starting from an open-source base model with limited initial SWE capability, SWE-Master demonstrates how systematical optimization method can elicit strong long-horizon SWE task solving abilities. We evaluate SWE-Master on SWE-bench Verified, a standard benchmark for realistic software engineering tasks. Under identical experimental settings, our approach achieves a resolve rate of 61.4\% with Qwen2.5-Coder-32B, substantially outperforming existing open-source baselines. By further incorporating test-time scaling~(TTS) with LLM-based environment feedback, SWE-Master reaches 70.8\% at TTS@8, demonstrating a strong performance potential. SWE-Master provides a practical and transparent foundation for advancing reproducible research on software engineering agents. The code is available at https://github.com/RUCAIBox/SWE-Master. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_03411 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | SWE-Master: Unleashing the Potential of Software Engineering Agents via Post-Training Song, Huatong Huang, Lisheng Sun, Shuang Jiang, Jinhao Le, Ran Cheng, Daixuan Chen, Guoxin Hu, Yiwen Chen, Zongchao Jia, Yiming Zhao, Wayne Xin Song, Yang Zhang, Tao Wen, Ji-Rong Software Engineering Computation and Language In this technical report, we present SWE-Master, an open-source and fully reproducible post-training framework for building effective software engineering agents. SWE-Master systematically explores the complete agent development pipeline, including teacher-trajectory synthesis and data curation, long-horizon SFT, RL with real execution feedback, and inference framework design. Starting from an open-source base model with limited initial SWE capability, SWE-Master demonstrates how systematical optimization method can elicit strong long-horizon SWE task solving abilities. We evaluate SWE-Master on SWE-bench Verified, a standard benchmark for realistic software engineering tasks. Under identical experimental settings, our approach achieves a resolve rate of 61.4\% with Qwen2.5-Coder-32B, substantially outperforming existing open-source baselines. By further incorporating test-time scaling~(TTS) with LLM-based environment feedback, SWE-Master reaches 70.8\% at TTS@8, demonstrating a strong performance potential. SWE-Master provides a practical and transparent foundation for advancing reproducible research on software engineering agents. The code is available at https://github.com/RUCAIBox/SWE-Master. |
| title | SWE-Master: Unleashing the Potential of Software Engineering Agents via Post-Training |
| topic | Software Engineering Computation and Language |
| url | https://arxiv.org/abs/2602.03411 |