EET: Experience-Driven Early Termination for Cost-Efficient Software Engineering Agents

TL;DR

EET is a method that reduces the cost of software engineering agents powered by large language models by guiding early termination based on learned experience, maintaining high task success rates.

Contribution

The paper introduces EET, a novel experience-driven early termination technique that significantly cuts costs in SE agents without sacrificing performance.

Findings

EET reduces total cost by 19%-55%, averaging 32%.

EET identifies early-termination opportunities for 11% of issues.

API calls, input tokens, and output tokens are reduced by 21%, 30%, and 25%, respectively.

Abstract

Software engineering (SE) agents powered by large language models are increasingly adopted in practice, yet they often incur substantial monetary cost. We introduce EET, an experience-driven early termination approach that reduces the cost of SE agents while preserving task performance. EET extracts structured experience from prior issue-resolution executions and leverages it to guide early termination during patch generation and selection, reducing unproductive iterations. We evaluate EET on the SWE-bench Verified benchmark across three representative SE agents. EET consistently reduces total cost by 19%-55% (32% on average), with negligible loss in resolution rate (at most 0.2%). These efficiency gains are achieved, on average, by identifying early-termination opportunities for 11% of issues and reducing API calls, input tokens, and output tokens by 21%, 30%, and 25%, respectively. We release the code, prompts, and data at https://github.com/IanWalls/EET.