SWE-Exp: Experience-Driven Software Issue Resolution
This addresses the inefficiency of redundant exploration in automated software engineering for developers, representing a new paradigm rather than an incremental improvement.
The paper tackled the problem of memoryless LLM agents in software issue resolution by introducing SWE-Exp, an experience-enhanced approach that distills knowledge from prior trajectories, achieving a state-of-the-art resolution rate of 41.6% Pass@1 on SWE-bench-Verified.
Recent advances in large language model (LLM) agents have shown remarkable progress in software issue resolution, leveraging advanced techniques such as multi-agent collaboration and Monte Carlo Tree Search (MCTS). However, current agents act as memoryless explorers - treating each problem separately without retaining or reusing knowledge from previous repair experiences. This leads to redundant exploration of failed trajectories and missed chances to adapt successful issue resolution methods to similar problems. To address this problem, we introduce SWE-Exp, an experience - enhanced approach that distills concise and actionable experience from prior agent trajectories, enabling continuous learning across issues. Our method introduces a multi-faceted experience bank that captures both successful and failed repair attempts. Specifically, it extracts reusable issue resolution knowledge at different levels - from high-level problem comprehension to specific code changes. Experiments show that SWE-Exp achieves state-of-the-art resolution rate (41.6% Pass@1) on SWE-bench-Verified under open-source agent frameworks. Our approach establishes a new paradigm in which automated software engineering agents systematically accumulate and leverage repair expertise, fundamentally shifting from trial-and-error exploration to strategic, experience-driven issue resolution.