Structurally Aligned Subtask-Level Memory for Software Engineering Agents
This addresses a granularity mismatch issue for autonomous software engineering agents, offering incremental improvements in long-horizon reasoning.
The paper tackles the problem of coarse instance-level memory in software engineering agents, which causes misguided retrieval for tasks with similar descriptions but distinct reasoning logic, by proposing a structurally aligned subtask-level memory method that improves mean Pass@1 by +4.7 percentage points on average over vanilla agents.
Large Language Models (LLMs) have demonstrated significant potential as autonomous software engineering (SWE) agents. Recent work has further explored augmenting these agents with memory mechanisms to support long-horizon reasoning. However, these approaches typically operate at a coarse instance granularity, treating the entire problem-solving episode as the atomic unit of storage and retrieval. We empirically demonstrate that instance-level memory suffers from a fundamental granularity mismatch, resulting in misguided retrieval when tasks with similar surface descriptions require distinct reasoning logic at specific stages. To address this, we propose Structurally Aligned Subtask-Level Memory, a method that aligns memory storage, retrieval, and updating with the agent's functional decomposition. Extensive experiments on SWE-bench Verified demonstrate that our method consistently outperforms both vanilla agents and strong instance-level memory baselines across diverse backbones, improving mean Pass@1 over the vanilla agent by +4.7 pp on average (e.g., +6.8 pp on Gemini 2.5 Pro). Performance gains grow with more interaction steps, showing that leveraging past experience benefits long-horizon reasoning in complex software engineering tasks.