PMCoder is presented, an issue-resolution agent that couples a hierarchical phase planner with episodic memory that outperforms either component alone and reduces repeated failed actions, empty-patch exits, and context-window exhaustion.
Abstract
Resolving a real software issue with a large language model (LLM) agent is a long repair episode, often tens to hundreds of steps spanning exploration, hypothesis, implementation, and verification. Success depends on both the base model's local reasoning and the agent's ability to maintain an evolving plan and remember observations across phases. Existing repository-level agents typically strengthen planning or memory in isolation, leaving long trajectories vulnerable to stale evidence, repeated failed edits, and verification inferred from the agent's own claims instead of execution evidence. We present PMCoder, an issue-resolution agent that couples a hierarchical phase planner with episodic memory. The coupling is bidirectional: the current plan phase conditions memory retrieval, while memory-derived trajectory statistics inform stuck detection and replanning. When available, issue-reproduction verdicts ground verification progress in execution evidence rather than self-reported completion. On SWE-bench Verified, PMCoder resolves an average of $25$ more cases ($+5.0$pp) than a harness-matched baseline, with gains persisting even where the reproduction gate never fires. Further Verified-500 evaluations show the same positive direction across Claude Haiku 4.5, DeepSeek-V4-Flash, and an OpenHands port, with at least $14$ additional resolved cases ($+2.8$pp). Separately, evaluation on TerminalWorld's official sample suggests that the plan-memory substrate transfers beyond issue reports. Ablation and trajectory analyses show where the gains come from: coupling planning and memory outperforms either component alone and reduces repeated failed actions, empty-patch exits, and context-window exhaustion.
Large language models for repository-level repair usually reset context for every issue, even though real maintenance often unfolds as a stream of tickets within the same codebase. We study this sequential regime on a stratified 8 × 25 subset of SWE-Bench Pro (eight projects, twenty-five instances each), running instances in commit-time order within each project so later repairs can draw on earlier trajectories. We present TwinMem-Agent, which pairs a test-patch-validate loop with episodic memory (grounded demonstrations) and semantic memory (LLM-distilled notes). After each issue, a memory module compacts the trajectory, merges updates under a fixed budget, and retrieves exemplars and guidelines at inference time, all without fine-tuning model weights. Using DeepSeek-V3.2 as the shared backbone and identical Dockerized instances, we compare TwinMem-Agent to mini-swe-agent on Pass@1 and average inference cost, and ablate the memory module on 25 Ansible instances measuring Pass@1 and RSR. TwinMem-Agent achieves 89.1% Pass@1 versus 60.0% for the baseline at roughly 18× lower cost ($0.0044 vs. $0.0798). Removing the memory module reduces Pass@1 by 8 pp and RSR by 12 pp, with the larger RSR drop confirming that accumulated cross-issue experience most directly benefits bug reproduction and cascades into higher patch success.
Benzhao Wu, Zhiwen Luo, Fangwen Mu et al.· SIGSOFT FSE Companion· 0 citations
Large Language Model (LLM)-based autonomous agents have shown significant promise in automating software engineering tasks, yet existing systems still suffer from two fundamental limitations: i) the lack of persistent experiential knowledge across debugging sessions, which forces agents to repeat exploratory mistakes, and ii) the static nature of prompt structures, which prevents agents from adapting their reasoning strategy to recurring failure modes. To address these gaps, we propose LS-CM (Long-Short Collaborative Memory), a dual-memory architecture that couples a Retrieval-Augmented Generation (RAG)-based short-term memory for code-base context with a Reinforcement Learning (RL)-driven long-term memory that distills strategic patterns from historical test failures and patch iterations. The core innovation is a self-evolving context engineering module that automatically refines the agent’s internal prompt structure based on execution feedback signals, treating the prompt as a configurable policy object that a gradient-free controller selects among pre-generated variants, rather than a fixed artifact. We evaluate LS-CM on two industry-standard benchmarks: SWE-bench, for end-to-end issue resolution on real-world GitHub repositories, and Defects4J, for bug localization and program repair in Java projects. Across multiple LLM backbones, LS-CM resolves 33.7% of SWE-bench Lite issues, a modest gain over the strongest reproduced baseline (Agentless, 32.0%), and improves Top-1 fault localization accuracy on Defects4J from 42.6% to 47.6% while reducing editing churn by 31%. We find that LS-CM’s most robust advantages lie in debugging efficiency, cross-episode strategy transfer, and run-to-run stability rather than in a single headline resolved-rate number. Ablation studies confirm that short-term retrieval, long-term policy learning, and self-evolving prompts each contribute to the gains. Our work demonstrates that explicitly modeling memory and prompt adaptation as first-class learning objects substantially enhances agent productivity on real-world software engineering tasks.
Tianjun Mo, Changhao Zhang, Jitong Zou et al.· IEEE Access· 0 citations
LLM agents that repair failures often discard successful corrections, forcing later episodes to rediscover similar solutions. We study whether finalized repair outcomes can improve subsequent Text-to-SQL episodes without parameter updates. We introduce MERIT, a training-free agent that maintains an online dual-polarity memory of oracle-verified corrections and observed unsuccessful directions. Under oracle-assisted benchmark feedback, only memories from earlier finalized episodes are eligible for retrieval. A deterministic classifier assigns a coarse failure type, which conditions a hybrid lexical-dense retriever before the frozen model generates each revision. Using Qwen2.5-7B-Instruct with identical initial predictions and repair budgets, MERIT improves execution accuracy over stateless iterative repair from \(66.34\%\) to \(69.79\%\) on Spider and from \(47.35\%\) to \(48.44\%\) on BIRD. Paired analyses provide clear evidence for the Spider gain but weaker evidence on BIRD. MERIT is not reliably separated from untyped dynamic retrieval on either benchmark, while Reflexion-style memory reaches \(51.24\%\) on BIRD at substantially higher inference cost. Ablations show that negative memory contributes modestly, the value of type conditioning and lexical--dense ranking is dataset dependent, and schema-local experience provides the most consistent benefit. These results clarify when causal cross-query memory improves repair and when broader memory representations remain preferable.
K. Vo, Tam Minh Chu, Anh Trac Duc Dinh et al.· 0 citations
Mimir is introduced, a neuro-symbolic memory that separates world memory from task memory and dynamically grounds them before each action, substantially outperforming current closed-source models.
Haoming Xu, Zhenlin He, Hengyi Wang et al.· 0 citations
Large language model (LLM) agents are increasingly used for multi-step, stateful tool-use tasks, yet production reliability remains limited. Unlike static software repair, agent repair must recover dynamic trajectories whose early decisions can propagate into later errors and external state changes. Existing automatic remedies address only part of this problem: blind retry adds no diagnosis, outcome feedback says whether a run failed but not where or why, and self-reflection often lacks grounded evidence to prevent the same failure from recurring. We present AgentTether, a run-time repair framework that automates post-run diagnosis and guided recovery without modifying the underlying agent or environment. AgentTether abstracts each run into Transition Units, links them through a dependency-aware Critical Transition Graph, and localizes failure-critical subtrajectories by combining an offline normal-behavior model with a run-local graph detector. It then converts the localized cause into behavior-scoped guidance backed by cross-iteration Repair Memory, and can optionally apply guarded run-time intervention to keep the correction active during re-execution. The same design can be deployed as an offline diagnostic-and-guidance tool or as an online repair layer. We evaluate AgentTether on 261 tau-bench tasks across three domains with Qwen3.7-max, and test cross-model transfer on Banking with GPT-5.4. On the hardest Banking domain, AgentTether repairs 59.04% (49/83) of initially failed Qwen3.7-max tasks and 65.12% (56/86) of initially failed GPT-5.4 tasks. Overall, AgentTether improves repair effectiveness while reducing agent turns and end-to-end approach tokens, suggesting a practical reliability layer that can wrap existing agent deployments, reduce wasted re-execution, and improve recovery without retraining the agent.
Chenyu Zhao, Shenglin Zhang, Wenwei Gu et al.· 2 citations
This review connects ten historical cognitive architectures, eight language-agent runtime families, and forty-two mechanism-focused modern systems to contribute a distinctive-mechanism catalog, an auditable evidence-depth framework, and a falsifiable agenda for testing these bundles as composable runtime invariants.