[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-86294-en":3,"doc-seo-86294-105":30,"detail-sidebar-cat-0-en-105":92},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},86294,13056703020460,"Valentina","https://ap-avatar.wpscdn.com/avatar/be000253dac470eee5d?_k=1778207105932848923",8,"Research & Report","RAGU: A Multi-Step GraphRAG Engine with a Compact Domain-Adapted LLM","Graph retrieval-augmented generation (GraphRAG) strengthens large language models with structured knowledge, but common systems build knowledge graphs in a single extraction pass, creating noisy entities and fragile retrieval. RAGU is an open-source modular GraphRAG engine that separates extraction from consolidation via two-stage typed extraction, DBSCAN-backed deduplication, LLM summarization, and Leiden community detection. A compact extractor is enabled by a language-skill hypothesis, leading to the 7B Meno-Lite-0.1 model and improved retrieval, synthesis, and single-GPU deployability under MIT.","RAGU: A Multi-Step GraphRAG Engine with a Compact Domain-Adapted LLM  \nMikhail Komarov 1 ,∗ Ivan Bondarenko2 ,∗ Stanislav Shtuka 1 ,3 Oleg Sedukhin†  \nRoman Shuvalov 1 Yana Dementyeva2 Matvey Solovyov 1 Nikolay O. Nikitin 1 1ITMO University 2Novosibirsk State University 3Far Eastern Federal University †Independent Researcher ∗Equal contribution  \n[i.bondarenko@g.nsu.ru](i.bondarenko@g.nsu.ru)  \narXiv :2607 . 1 1683v 1 [ cs .CL] 13 Jul 2026  \nAbstract  \nGraph 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.  \nAccordingly, we train Meno-Lite-0.1, a 7 B model optimized for language skills, which outperforms Qwen2.5-32B on knowledgegraph 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. ≤0.76) and overtakes HippoRAG 2 on synthesis tasks; on multi-hop factoid QA, the apparent HippoRAG 2 advantage is shown to be largely an answer-format artifact. RAGU is installable via pip install graph_ragu, runs on a single GPU, and is released under MIT license. The source code is publicly available at [https:](https:)//[github.com/RaguTeam/RAGU](github.com/RaguTeam/RAGU), and the Meno-Lite-0.1 model can be obtained from  \n[https://huggingface.co/bond005/](https://huggingface.co/bond005/)[ ](https://huggingface.co/bond005/)[meno-lite-0.1](meno-lite-0.1.)[.](meno-lite-0.1.)  \n1 Introduction  \nRetrieval-augmented generation (RAG) grounds large language models (LLMs) in external knowledge (Lewis et al., 2020 ; Gao et al., 2023) . Traditional RAG retrieves flat chunks without capturing cross-document entity relationships. Graph  \nRAG (GraphRAG) (Edge et al., 2024 ; Guo et al., 2025 ; Gutiérrez et al., 2025) addresses this by building a knowledge graph and using graph traversal during retrieval, but practical adoption faces three obstacles.  \nObstacle 1: Single-pass extraction. Current systems treat knowledge graph construction as a single LLM extraction pass, producing noisy, duplicated entities with no mechanism to consolidate information across chunks.  \nObstacle 2: Dependence on expensive LLMs.  \nExtraction quality determines graph quality, so practitioners default to large API models (GPT-4-class) . This rests on a false premise: the capabilities an LLM needs inside a RAG pipeline—comprehension, extraction, reasoning over context—are language skills, not factual recall. As we show next, language skills grow weakly with model size, while world knowledge scales steeply. A compact, skill-oriented model is therefore sufficient.  \nObstacle 3: Engineering immaturity. Many open-source frameworks suffer from installation failures or unsafe code paths (e.g., eval() on raw LLM output) . A GraphRAG engine should be both semantically strong and engineerable: installable, testable, and deployable on cost-effective hardware.  \nLanguage/World Knowledge Hypothesis. We hypothesize that world knowledge scales nearlinearly with parameter count, whereas language skills scale markedly more slowly. Figure 1 tests this on the Qwen2.5-Instruct family (Yang et al., 2024): on CheGeKa (Taktasheva et al., 2022)(world-knowledge quiz), F1 increases 21.1 × from 0.5 B to 72 B, while on MultiQ (all facts in-context), it increases only 4× . In a GraphRAG pipeline, the LLM extracts ent","cbCaigD8S6iJR9l2","https://ap.wps.com/l/cbCaigD8S6iJR9l2","pdf",560782,9,1,10,"English","en",105,"# Abstract\n# Introduction\n## Obstacles and language/world knowledge hypothesis\n## Proposed artifacts and system overview\n# System Description\n## Multi-Step Graph Construction\n## Two-stage extraction and consolidation","[{\"question\":\"What problem does RAGU address in GraphRAG knowledge graph construction?\",\"answer\":\"RAGU targets noise and brittleness caused by single-pass extraction, which produces duplicated or poorly consolidated entities without mechanisms to merge information across chunks.\"},{\"question\":\"How does RAGU differ from prior GraphRAG systems like Microsoft GraphRAG and LightRAG?\",\"answer\":\"RAGU introduces an explicit multi-step consolidation stage, separating extraction from consolidation and producing cleaner, more connected knowledge graphs.\"},{\"question\":\"What is the role of the compact domain-adapted LLM (Meno-Lite-0.1) in RAGU?\",\"answer\":\"Meno-Lite-0.1 is trained to support language-oriented skills needed inside the pipeline (comprehension, extraction, and contextual reasoning), enabling strong knowledge-graph construction with a 7B model and cost-effective single-GPU running.\"}]",1784210168,25,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":87,"head_meta":89,"extra_data":91,"updated_unix":28},"ragu-a-multi-step-graphrag-engine-with-a-compact-domain-adapted-llm","",{"@graph":36,"@context":86},[37,54,69],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/ragu-a-multi-step-graphrag-engine-with-a-compact-domain-adapted-llm/86294/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":24,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-07-27","2026-07-16",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"What problem does RAGU address in GraphRAG knowledge graph construction?","Question",{"text":76,"@type":77},"RAGU targets noise and brittleness caused by single-pass extraction, which produces duplicated or poorly consolidated entities without mechanisms to merge information across chunks.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"How does RAGU differ from prior GraphRAG systems like Microsoft GraphRAG and LightRAG?",{"text":81,"@type":77},"RAGU introduces an explicit multi-step consolidation stage, separating extraction from consolidation and producing cleaner, more connected knowledge graphs.",{"name":83,"@type":74,"acceptedAnswer":84},"What is the role of the compact domain-adapted LLM (Meno-Lite-0.1) in RAGU?",{"text":85,"@type":77},"Meno-Lite-0.1 is trained to support language-oriented skills needed inside the pipeline (comprehension, extraction, and contextual reasoning), enabling strong knowledge-graph construction with a 7B model and cost-effective single-GPU running.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,121,124,128,131,134],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":46,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":46,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":20,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":22,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":22,"slug":133},"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":107,"slug":137},19,"General","general"]