arXiv cs.AI· Songwei Yu, Ziqi Miao, Jingxuan Xu, Junhao Fan, Haoyang Huang, Yuan Xiong, Zihang Xu, Weihao Li, Wenqiang Zhu, Yuechen Luo, Xinshun Feng, Yanan Wu, Ken Deng·· 2 天前AI 评分34
SWE-Nexus：从测试执行自动组合软件工程任务
SWE-Nexus: Automated Composition of Software Engineering Tasks from Test Execution
AI 导读
SWE-Nexus 是一个全自动流水线，通过测试执行构建仓库级实现任务，由四个构建智能体筛选测试、适配追踪与评估协议并生成行为规范。其基准包含来自 105 个仓库、六种编程语言的 600 个任务（116 个原子任务、484 个组合任务），在三种智能体脚手架下评测 12 个模型，最佳配置单次尝试解决 37.8% 的任务。
正文 · 原文
Authors:Songwei Yu, Ziqi Miao, Jingxuan Xu, Junhao Fan, Haoyang Huang, Yuan Xiong, Zihang Xu, Weihao Li, Wenqiang Zhu, Yuechen Luo, Xinshun Feng, Yanan Wu, Ken Deng
Abstract:Historical code changes provide convenient task boundaries, but do not identify which production functions selected tests exercise or support composing validated behaviors beyond a single change. We present SWE-Nexus, a fully automated pipeline that constructs repository-level implementation tasks from test execution. Four construction agents select tests, adapt tracing and evaluation protocols to each repository, and generate behavioral specifications. For each selected test file, tracing records an observed execution footprint of production function bodies. A scope is retained only if all expected scoring points pass in the original repository and masking the traced bodies causes at least one failure or error. Compatible scopes are composed deterministically, with shared functions counted once and every constituent test file revalidated under the complete combined mask. The benchmark contains 600 tasks from 105 repositories across six programming languages, including 116 atomic and 484 composite tasks. We evaluate 12 models under three agent scaffolds. The best configuration resolves 37.8% of tasks in one attempt, and scope-weighted scoring measures partial recovery. SWE-Nexus enables task construction from validated behavior beyond individual historical changes.
| Subjects: | Software Engineering (cs.SE); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.20038 [cs.SE] |
| (or arXiv:2610.20038v2 [cs.SE] for this version) | |
| https://doi.org/10.48550/arXiv.2610.20038 arXiv-issued DOI via DataCite |
Submission history
From: Miao Ziqi [view email]
[v1]
Wed, 14 Oct 2026 11:45:59 UTC (626 KB)
[v2]
Thu, 15 Oct 2026 14:01:09 UTC (626 KB)
来源：arXiv cs.AI · arxiv.org