Reproduce “Coupling Planning with Episodic Memory in LLM Agents for Software Issue Resolution” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Jiahao Zhang, Yifan Zhang, Yu Huang
PMCoder coordinates a repair plan with episodic memory for repository issue resolution. It tracks repair phases and uses the active subtask to retrieve notes, while memory-derived signs of repeated trouble can prompt replanning. On SWE-bench Verified, it averaged 167.3/500 resolved issues (33.5%) over three runs, versus 142.3/500 (28.5%) for baseline; factorial ablation supports plan–memory interaction, without isolating grounding or recovery.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.