SWE-Gate Benchmark for Software Agents
SWE-Gate: Passing Functional Tests Is Not Enough for Software Engineering Agents
SWE-Gate addresses critical evaluation gaps in coding agents, which currently only test functional correctness but ignore real-world review constraints that determine patch acceptance.
SWE-Gate introduces a repository-level benchmark for software engineering agents that evaluates both functional correctness and review constraint compliance. The benchmark contains 303 repair instances across 75 open-source Python repositories. Experiments with four LLM backends show that 221 out of 644 repairs passing functional tests fail to satisfy review constraints, revealing a significant gap in evaluation metrics.
SWE-Gate: Passing Functional Tests Is Not Enough for Software Engineering Agents
Repository-level software engineering benchmarks have significantly advanced the evaluation of coding agents, but existing benchmarks primarily measure whether generated patches pass functional tests and overlook review-derived acceptance constraints (review constraints) that often influence whether a patch is acceptable in real-world software development. We introduce SWE-Gate, a repository-level benchmark for software engineering agents that explicitly evaluates review constraint compliance alongside functional correctness. SWE-Gate derives review constraints from real pull request review comments and synthesizes repository-level repair instances around these constraints. Each instance provides separate functional and constraint tests, together with non-compliant and gold patches, enabling explicit separation between issue resolution capability and review constraint compliance. We construct SWE-Gate with 303 repository-level repair instances spanning 75 open-source Python repositories across diverse software domains. Experiments with four LLM backends spanning different capability levels under a common coding-agent scaffold reveal a substantial gap between functional success and success under the complete repair specification: among 644 repairs that pass the functional tests, 221 fail to satisfy the provided review constraints. These findings show that functional-only evaluation overestimates agents' ability to satisfy the full requirements of repository-level repair tasks. The replication package including code, data, and experimental results is available at https://github.com/DeepSoftwareAnalytics/SWE-Gate.
