# Structurally Aligned Subtask-Level Memory for Software Engineering Agents

arXiv: https://arxiv.org/abs/2602.21611 · HTML: https://mlsanity.com/paper/2602.21611/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- Placed in Software engineering coding agents, one of 16 directions in LLM agents and tool use.
- Builds on 36 of its 48 references that are in the map.
- First posted on arXiv 25 Feb 2026.

## Placed in

- LLM agents and tool use / [Software engineering coding agents](https://mlsanity.com/agents/software-engineering-coding-agents/)

## Builds on (most cited first, 30 of 36)

- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Reflexion: Language Agents with Verbal Reinforcement Learning](https://mlsanity.com/paper/2303.11366/) (2023)
- [Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities](https://mlsanity.com/paper/2507.06261/) (2025)
- [Program Synthesis with Large Language Models](https://mlsanity.com/paper/2108.07732/) (2021)
- [Generative Agents: Interactive Simulacra of Human Behavior](https://mlsanity.com/paper/2304.03442/) (2023)
- [Lost in the Middle: How Language Models Use Long Contexts](https://mlsanity.com/paper/2307.03172/) (2023)
- [SWE-bench: Can Language Models Resolve Real-World GitHub Issues?](https://mlsanity.com/paper/2310.06770/) (2023)
- [SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering](https://mlsanity.com/paper/2405.15793/) (2024)
- [OpenHands: An Open Platform for AI Software Developers as Generalist Agents](https://mlsanity.com/paper/2407.16741/) (2024)
- [Measuring Coding Challenge Competence With APPS](https://mlsanity.com/paper/2105.09938/) (2021)
- [GLM-4.5: Agentic, Reasoning, and Coding (ARC) Foundation Models](https://mlsanity.com/paper/2508.06471/) (2025)
- [Cognitive Architectures for Language Agents](https://mlsanity.com/paper/2309.02427/) (2023)
- [Agentless: Demystifying LLM-based Software Engineering Agents](https://mlsanity.com/paper/2407.01489/) (2024)
- [RepoBench: Benchmarking Repository-Level Code Auto-Completion Systems](https://mlsanity.com/paper/2306.03091/) (2023)
- [Memory in the Age of AI Agents](https://mlsanity.com/paper/2512.13564/) (2025)
- [AutoCodeRover: Autonomous Program Improvement](https://mlsanity.com/paper/2404.05427/) (2024)
- [MiniMax-M1: Scaling Test-Time Compute Efficiently with Lightning Attention](https://mlsanity.com/paper/2506.13585/) (2025)
- [SWE-bench Multimodal: Do AI Systems Generalize to Visual Software Domains?](https://mlsanity.com/paper/2410.03859/) (2024)
- [Multi-SWE-bench: A Multilingual Benchmark for Issue Resolving](https://mlsanity.com/paper/2504.02605/) (2025)
- [SWE-Search: Enhancing Software Agents with Monte Carlo Tree Search and Iterative Refinement](https://mlsanity.com/paper/2410.20285/) (2024)
- [Context Length Alone Hurts LLM Performance Despite Perfect Retrieval](https://mlsanity.com/paper/2510.05381/) (2025)
- [How Memory Management Impacts LLM Agents: An Empirical Study of Experience-Following Behavior](https://mlsanity.com/paper/2505.16067/) (2025)
- [Live-SWE-agent: Can Software Engineering Agents Self-Evolve on the Fly?](https://mlsanity.com/paper/2511.13646/) (2025)
- [DevEval: A Manually-Annotated Code Generation Benchmark Aligned with Real-World Code Repositories](https://mlsanity.com/paper/2405.19856/) (2024)
- [SWE-Exp: Experience-Driven Software Issue Resolution](https://mlsanity.com/paper/2507.23361/) (2025)
- [FEA-Bench: A Benchmark for Evaluating Repository-Level Code Generation for Feature Implementation](https://mlsanity.com/paper/2503.06680/) (2025)
- [EXPEREPAIR: Dual-Memory Enhanced LLM-based Repository-Level Program Repair](https://mlsanity.com/paper/2506.10484/) (2025)
- [TRAD: Enhancing LLM Agents with Step-Wise Thought Retrieval and Aligned Decision](https://mlsanity.com/paper/2403.06221/) (2024)
- [Code Researcher: Deep Research Agent for Large Systems Code and Commit History](https://mlsanity.com/paper/2506.11060/) (2025)
- [Improving Code Localization with Repository Memory](https://mlsanity.com/paper/2510.01003/) (2025)

## Built on by (most cited first, 12 of 12)

- [CODESKILL: Learning Self-Evolving Skills for Coding Agents](https://mlsanity.com/paper/2605.25430/) (2026)
- [From Question Answering to Task Completion: A Survey on Agent System and Harness Design](https://mlsanity.com/paper/2606.20683/) (2026)
- [Learning to Commit: Generating Organic Pull Requests via Online Repository Memory](https://mlsanity.com/paper/2603.26664/) (2026)
- [Agent Memory Distillation: Empowering Small LLM Agents with Hierarchical Teacher Memory](https://mlsanity.com/paper/2608.07169/) (2026)
- [Self-Evolving Coding Agents](https://mlsanity.com/paper/2608.03392/) (2026)
- [Break It Down, Pass It On: Cross-Task Skill Transfer in LLM Agents](https://mlsanity.com/paper/2608.20274/) (2026)
- [ContextSniper: AntTrail's Token-Efficient Code Memory for Repository-Level Program Repair](https://mlsanity.com/paper/2607.01916/) (2026)
- [Coupling Planning with Episodic Memory in LLM Agents for Software Issue Resolution](https://mlsanity.com/paper/2608.06811/) (2026)
- [Enhancing Software Engineering Through Closed-Loop Memory Optimization](https://mlsanity.com/paper/2606.05646/) (2026)
- [DreamBench-SWE: A Multi-Session Memory-Hygiene Benchmark for Software Agents](https://mlsanity.com/paper/2608.20664/) (2026)
- [The Missing Complement: State-Conditioned Minimal Sufficient Evidence for Coding Agents](https://mlsanity.com/paper/2609.20050/) (2026)
- [AdaRepair-Mem: Adaptive Experience Orchestration for Repository-Level Program Repair](https://mlsanity.com/paper/2609.20130/) (2026)

## Abstract (arXiv)

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.