跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2608.18389 · v2

Programmable Distributions of Semantics-Preserving Variants for Evaluating Code Agents

首次提交 2026-08-19 07:46
版本更新 2026-11-13 06:58
雷达观测 2026-11-17 04:58
arXiv ↗PDF ↗

摘要

Code agents are increasingly deployed to resolve real software issues, yet their reliability under superficial code variations remains poorly understood. We evaluate whether coding agents that repair repository-level issues remain reliable when the surrounding codebase is rewritten into a semantically equivalent form. We sample semantically equivalent variants of a repository from an explicitly defined distribution and compare the agent's resolve rate on them with that on the original. Different distributions reveal different aspects of an agent, such as its sensitivity to routine refactoring or to misleading code near the location of a fix. We develop VDL, a probabilistic programming language for expressing such distributions, whose programs randomly apply semantics-preserving transformations (SPTs) to a repository. For our evaluation, we use it to define a generic distribution that is not targeted at any particular failure mode. We evaluate two agentic scaffolds (mini-SWE agent and OpenCode), each backed by one of four frontier models (Claude Opus 4.5, Kimi K2.5, MiniMax M2.5, and Qwen 3.6-27B), on instances from SWE-bench Verified and SWE-bench Pro. Resolve rates drop by up to 6.7 percentage points, significantly in 6 of 16 configurations. No ranking of models by robustness holds across scaffolds, revealing a jagged robustness frontier.

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存,登录后共享已有结果。

版本差异

本站首次收录为 v2,暂无更早版本可对照。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v2 · 2026-11-13 06:58 · 最新观测版本

Programmable Distributions of Semantics-Preserving Variants for Evaluating Code Agents

Code agents are increasingly deployed to resolve real software issues, yet their reliability under superficial code variations remains poorly understood. We evaluate whether coding agents that repair repository-level issues remain reliable when the surrounding codebase is rewritten into a semantically equivalent form. We sample semantically equivalent variants of a repository from an explicitly defined distribution and compare the agent's resolve rate on them with that on the original. Different distributions reveal different aspects of an agent, such as its sensitivity to routine refactoring or to misleading code near the location of a fix. We develop VDL, a probabilistic programming language for expressing such distributions, whose programs randomly apply semantics-preserving transformations (SPTs) to a repository. For our evaluation, we use it to define a generic distribution that is not targeted at any particular failure mode. We evaluate two agentic scaffolds (mini-SWE agent and OpenCode), each backed by one of four frontier models (Claude Opus 4.5, Kimi K2.5, MiniMax M2.5, and Qwen 3.6-27B), on instances from SWE-bench Verified and SWE-bench Pro. Resolve rates drop by up to 6.7 percentage points, significantly in 6 of 16 configurations. No ranking of models by robustness holds across scaffolds, revealing a jagged robustness frontier.

查看此版本解读 →查看此版本原文 ↗