AI 中文总结
ECLoop是介于编码智能体与仓库间的证据条件执行层，可防止过早提交，在SWE-bench Verified上提升Pass@1指标4.8-11.8个百分点，同时降低平均token消耗。
AI 中文摘要
基于大语言模型（LLM）的编码智能体常未检查足够的仓库证据就编辑源代码或提交补丁，这种失败模式被称为过早提交。本文提出ECLoop，一种介于智能体与仓库之间的执行层，用于强制实施证据条件执行。对于每个任务，ECLoop利用问题描述和仓库结构，编译一组条件，规定智能体在进行各类代码修改或补丁提交前应观察到的内容。执行过程中，ECLoop会跟踪智能体运行轨迹已满足的条件，推迟所需条件未达成的拟执行操作。在SWE-bench Verified的全部500个实例上，采用两种语言模型和两种智能体框架进行评估，结果显示ECLoop可提升Pass@1指标4.8至11.8个百分点，且无需重新训练模型或修改框架。消融实验表明，ECLoop的三项操作各有独特价值，且结构化证据条件的效果优于同等长度的自然语言摘要。这些提升未带来额外推理成本：通过在智能体采取无支持操作前进行重定向，ECLoop可将平均token消耗降低最多12.1%。
英文摘要
LLM-based coding agents often edit source code or submit patches before examining enough repository evidence to justify the change, a failure pattern we call premature commitment. We present ECLoop, an execution layer that interposes between the agent and the repository to enforce evidence-conditioned execution. For each task, ECLoop uses the issue description and repository structure to compile a set of conditions specifying what the agent should observe before each type of code modification or patch submission. During execution, ECLoop tracks which conditions the agent's runtime trajectory has satisfied and postpones any proposed action whose required conditions remain unmet. Evaluated on all 500 instances of SWE-bench Verified with two language models and two agent scaffolds, ECLoop raises Pass@1 by 4.8-11.8 percentage points without model retraining or scaffold changes. Ablation experiments show that each of ECLoop's three operations contributes distinct value and that structured evidence conditions outperform an equivalent natural-language summary. These gains come at no additional inference cost: by redirecting the agent before it pursues unsupported actions, ECLoop lowers average token consumption by up to 12.1%.
Comments9 pages
