发表机构
ITMO University; Novosibirsk State University; Far Eastern Federal University（圣彼得堡国立信息技术机械与光学大学; 新西伯利亚国立大学; 远东联邦大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究针对现有GraphRAG系统问题，提出RAGU引擎，通过分离提取与合并多步处理知识图。核心方法包括两阶段提取等。贡献在于训练的Meno-Lite-0.1模型表现出色，在多方面超越其他模型，且RAGU安装简单、运行要求低并开源。
AI 中文摘要
图检索增强生成（GraphRAG）通过结构化知识增强大语言模型，但现有系统在单次提取过程中构建知识图，产生有噪声的实体和脆弱的检索。开源模块化GraphRAG引擎RAGU通过将提取与合并分离来解决此问题：实体和关系经过两阶段类型提取、基于DBSCAN的去重、语言模型总结和莱顿社区检测。关键见解促使开发紧凑提取器：流水线中的语言模型所需技能（理解、提取、上下文推理）是随模型大小增长微弱的语言技能，与事实世界知识不同。因此，训练了针对语言技能优化的7B模型Meno-Lite-0.1，在知识图构建方面优于Qwen2.5-32B（相对调和均值提高12.5%），在英语GraphRAG任务上与之匹配。在GraphRAG-Bench（医学）上，RAGU在每个事实层面检索到最完整的上下文（证据召回率高达0.84，而其他模型≤0.76），在合成任务上超过HippoRAG2；在多跳事实问答中，HippoRAG2的明显优势很大程度上是答案格式造成的。RAGU可通过pip install graph_ragu安装，在单个GPU上运行，遵循MIT许可发布。
英文摘要
Graph retrieval-augmented generation (GraphRAG) enhances large language models with structured knowledge, yet existing systems construct knowledge graphs in a single extraction pass, producing noisy entities and brittle retrieval. RAGU, an open-source modular GraphRAG engine, addresses this by separating extraction from consolidation: entities and relations pass through two-stage typed extraction, DBSCAN-backed deduplication, LLM summarization, and Leiden community detection. A key insight motivates a compact extractor: the skills an in-pipeline LLM needs - comprehension, extraction, reasoning over context - are language skills that grow only weakly with model size, unlike factual world knowledge. Accordingly, we train Meno-Lite-0.1, a 7B model optimized for language skills, which outperforms Qwen2.5-32B on knowledge-graph construction (+12.5% relative harmonic mean) and matches it on English GraphRAG tasks. On GraphRAG-Bench (Medical), RAGU retrieves the most complete context at every factoid level (evidence recall up to 0.84 vs. $\leq$0.76) and overtakes HippoRAG2 on synthesis tasks; on multi-hop factoid QA, the apparent HippoRAG2 advantage is shown to be largely an answer-format artifact. RAGU is installable via $\texttt{pip install graph_ragu}$, runs on a single GPU, and is released under MIT. The source code is publicly available at https://github.com/RaguTeam/RAGU, and the Meno-Lite-0.1 model can be obtained from https://huggingface.co/bond005/meno-lite-0.1.
