一句话：提出多解推理与合成框架 Prism，通过全局/局部探索与多智能体合成提升仓库级问题修复的通过率。
基于大语言模型（LLM）的智能体在仓库级问题修复上已有进展，但现有框架面临两个瓶颈：候选修复方案的多样性不足，以及难以融合各候选方案的互补优势。为此，作者提出 Prism，一个采用由粗到细范式的多解推理与合成框架，包含三个阶段：（1）带对比语义约束的全局探索机制，通过鼓励历史解之间的分歧来扩大搜索空间；（2）受 A* 算法启发的局部探索机制，定位合适的分支点以生成有差异的子解；（3）多智能体协同的解合成过程，由 reviewer 与 judge 智能体共同评估、互补并整合部分正确的候选解。在 SWE-bench Verified 与 SWE-bench-Live Lite 上评估：配合 Claude-4.6-Sonnet 与 DeepSeek-V3.2-Reasoner，在 SWE-bench Verified 上的 Pass@1 分别达到 82.0% 与 80.0%，优于对比基线；在旨在缓解数据泄漏的 SWE-bench-Live Lite 上也优于所报告的基线。分析表明 Prism 扩展了解空间，其解合成阶段独特解决了 15 个孤立候选解未能处理的问题。
Large language model (LLM)-based agents have achieved notable success in repository-level issue resolution, yet existing frameworks still face two key bottlenecks: insufficient diversity among candidate repair solutions and limited ability to combine complementary strengths across candidate solutions. To address these challenges, we introduce Prism, a multi-solution reasoning and synthesis framework that adopts a coarse-to-fine paradigm to systematically generate, refine, and integrate diverse repair solutions. Prism comprises three stages: (1) a global exploration mechanism with contrastive semantic constraints that broadens the search space by encouraging divergence across historical solutions; (2) an A* algorithm-inspired local exploration mechanism that identifies suitable branching points to generate divergent sub-solutions; and (3) a multi-agent collaborative solution synthesis process where reviewer and judge agents jointly evaluate, complement, and integrate partially correct candidate solutions. We evaluate Prism on SWE-bench Verified and SWE-bench-Live Lite. With Claude-4.6-Sonnet and DeepSeek-V3.2-Reasoner, Prism achieves Pass@1 rates of 82.0% and 80.0% on SWE-bench Verified, respectively, outperforming the compared baselines. On SWE-bench-Live Lite, designed to reduce data leakage concerns, Prism outperforms the reported baselines. Further analyses show that Prism expands the solution space and that its solution synthesis stage uniquely resolves 15 issues unaddressed by isolated candidates.
@inproceedings{paperbot12275,
title = {Prism: A Multi-solution Reasoning and Synthesis Framework for Repository-Level Issue Resolution},
author = {Yibo Wang and Guotian Wang and Zhihao Peng and Ying Wang and Chang Xu and Zhiliang Zhu},
booktitle = {ASE},
year = {2026},
doi = {10.1145/3832783.3837451}
}