Know Before Fix: QA-Driven Repository Knowledge Acquisition for Software Issue Resolution

2026-07-13

AI Summary

This work addresses the challenge that large language models (LLMs) often introduce factual errors in automated program repair due to insufficient understanding of the codebase. To mitigate this, the authors propose ACQUIRE, a framework that explicitly acquires codebase knowledge through a question-answering mechanism. ACQUIRE first employs a collaborative process between a questioner and an answerer to generate structured, evidence-backed question-answer pairs; a separate repair module then leverages this curated knowledge to produce patches. By decoupling knowledge acquisition from patch generation, ACQUIRE emulates the developer practice of “understand first, fix later,” transforming implicit knowledge gaps into explicit, reliable facts. Evaluated on the SWE-bench Verified benchmark, ACQUIRE substantially outperforms existing pre-exploration approaches, achieving up to a 4.4 percentage point improvement in Pass@1 while maintaining manageable computational overhead.

Abstract

LLM-based coding agents have significantly advanced automated software issue resolution, yet they remain highly prone to factual errors caused by insufficient repository understanding. Recent methods attempt to mitigate this limitation through pre-repair repository exploration; however, their fix-driven strategies explore repositories without identifying the agent's knowledge gaps, often yielding imprecise context that fails to bridge the underlying understanding deficit. In this paper, we propose ACQUIRE, a QA-driven framework for software issue resolution. Mirroring how experienced developers first comprehend unfamiliar code before attempting a fix, ACQUIRE explicitly acquires repository knowledge prior to repair. The framework decouples knowledge acquisition from patch generation through two stages: in the first stage, a Questioner and an Answerer collaborate to acquire structured repository knowledge, where the Questioner poses targeted questions and the Answerer produces evidence-grounded answers through autonomous exploration; in the second stage, the Resolver leverages the resulting QA knowledge to generate informed patches. By transforming implicit knowledge gaps into explicit, factually reliable understanding, ACQUIRE accelerates knowledge-intensive repair stages and enables more accurate resolution. Experiments on SWE-bench Verified demonstrate that ACQUIRE consistently outperforms representative pre-repair methods, raising Pass@1 by up to 4.4 percentage points with modest additional cost and time.

Problem

• repository understanding
• factual errors
• knowledge gaps
• software issue resolution
• code comprehension

Innovation

• QA-driven
• repository knowledge acquisition
• LLM-based code repair
• knowledge gap identification
• structured QA framework