7 GraphRAG designs that outperform keyword search. Fun, quick insights into using LLMs and reducing hallucinations with relationships medium.com/@ThinkingLoo...
7 Graph RAG Designs That Outperform Keyword Search
Practical patterns that fuse knowledge graphs with retrieval-augmented generation to surface context, reduce hallucinations, and answer…
medium.com
November 11, 2025 at 12:15 PM
7 GraphRAG designs that outperform keyword search. Fun, quick insights into using LLMs and reducing hallucinations with relationships medium.com/@ThinkingLoo...
<a href="https://speakerdeck.com/negi111111/pythondegou-zhu-suruquan-guo-shi-ting-cun-naretuzigurahu-graphragwoyong-itayi-wei-de-di-yu-jian-suo-henoying-yong" class="hover:underline text-blue-600 dark:text-sky-400 no-card-link" target="_blank" rel="noopener" data-link="bsky">speakerdeck.com/negi111111...
Pythonで構築する全国市町村ナレッジグラフ: GraphRAGを用いた意味的地域検索への応用
Pythonで構築する全国市町村ナレッジグラフ: GraphRAGを用いた意味的地域検索への応用
Pythonで構築する全国市町村ナレッジグラフ: GraphRAGを用いた意味的地域検索への応用
speakerdeck.com
November 11, 2025 at 4:04 AM
<a href="https://speakerdeck.com/negi111111/pythondegou-zhu-suruquan-guo-shi-ting-cun-naretuzigurahu-graphragwoyong-itayi-wei-de-di-yu-jian-suo-henoying-yong" class="hover:underline text-blue-600 dark:text-sky-400 no-card-link" target="_blank" rel="noopener" data-link="bsky">speakerdeck.com/negi111111...
Pythonで構築する全国市町村ナレッジグラフ: GraphRAGを用いた意味的地域検索への応用
Pythonで構築する全国市町村ナレッジグラフ: GraphRAGを用いた意味的地域検索への応用
Pythonで構築する全国市町村ナレッジグラフ: GraphRAGを用いた意味的地域検索への応用
https://speakerdeck.com/negi111111/pythondegou-zhu-suruquan-guo-shi-ting-cun-naretuzigurahu-graphragwoyong-itayi-wei-de-di-yu-jian-suo-henoying-yong
https://speakerdeck.com/negi111111/pythondegou-zhu-suruquan-guo-shi-ting-cun-naretuzigurahu-graphragwoyong-itayi-wei-de-di-yu-jian-suo-henoying-yong
Pythonで構築する全国市町村ナレッジグラフ: GraphRAGを用いた意味的地域検索への応用
PyCon mini 東海 2025で登壇した内容に関する資料です。
URL：https://tokai.pycon.jp/2025/
内容：
全国1741市町村の統計データであるSSDSE（教育用標準データセット）と住民基本台帳人口移動データを用いた市町村間の人の流れを示す人口移動データを組み合…
speakerdeck.com
November 10, 2025 at 3:28 AM
Pythonで構築する全国市町村ナレッジグラフ: GraphRAGを用いた意味的地域検索への応用
https://speakerdeck.com/negi111111/pythondegou-zhu-suruquan-guo-shi-ting-cun-naretuzigurahu-graphragwoyong-itayi-wei-de-di-yu-jian-suo-henoying-yong
https://speakerdeck.com/negi111111/pythondegou-zhu-suruquan-guo-shi-ting-cun-naretuzigurahu-graphragwoyong-itayi-wei-de-di-yu-jian-suo-henoying-yong
Platform engineering is at a breaking point. Traditional tools can’t handle modern infrastructure complexity.
CAIPE and AGNTCY are solving this with GraphRAG, a system combining knowledge graphs and LLMs to help teams manage structured data and scale operations.
cs.co/633247MSK8
CAIPE and AGNTCY are solving this with GraphRAG, a system combining knowledge graphs and LLMs to help teams manage structured data and scale operations.
cs.co/633247MSK8
November 5, 2025 at 3:15 PM
Platform engineering is at a breaking point. Traditional tools can’t handle modern infrastructure complexity.
CAIPE and AGNTCY are solving this with GraphRAG, a system combining knowledge graphs and LLMs to help teams manage structured data and scale operations.
cs.co/633247MSK8
CAIPE and AGNTCY are solving this with GraphRAG, a system combining knowledge graphs and LLMs to help teams manage structured data and scale operations.
cs.co/633247MSK8
Just read “HubLink: A Novel QA Retrieval over Knowledge Graphs”, Best Paper at #RAGEKG 2025 🧠
Instead of NL→SPARQL, it slices KG into hubs, embeds those subgraphs, does GraphRAG, and returns the triples used for provenance.
ceur-ws.org/Vol-4079
#KnowledgeGraphs #RDF
Instead of NL→SPARQL, it slices KG into hubs, embeds those subgraphs, does GraphRAG, and returns the triples used for provenance.
ceur-ws.org/Vol-4079
#KnowledgeGraphs #RDF
November 5, 2025 at 1:30 PM
Just read “HubLink: A Novel QA Retrieval over Knowledge Graphs”, Best Paper at #RAGEKG 2025 🧠
Instead of NL→SPARQL, it slices KG into hubs, embeds those subgraphs, does GraphRAG, and returns the triples used for provenance.
ceur-ws.org/Vol-4079
#KnowledgeGraphs #RDF
Instead of NL→SPARQL, it slices KG into hubs, embeds those subgraphs, does GraphRAG, and returns the triples used for provenance.
ceur-ws.org/Vol-4079
#KnowledgeGraphs #RDF
Tejas Sarnaik, Manan Shah, Ravi Hegde
PROPEX-RAG: Enhanced GraphRAG using Prompt-Driven Prompt Execution
https://arxiv.org/abs/2511.01802
PROPEX-RAG: Enhanced GraphRAG using Prompt-Driven Prompt Execution
https://arxiv.org/abs/2511.01802
November 4, 2025 at 6:39 AM
Tejas Sarnaik, Manan Shah, Ravi Hegde
PROPEX-RAG: Enhanced GraphRAG using Prompt-Driven Prompt Execution
https://arxiv.org/abs/2511.01802
PROPEX-RAG: Enhanced GraphRAG using Prompt-Driven Prompt Execution
https://arxiv.org/abs/2511.01802
Tejas Sarnaik, Manan Shah, Ravi Hegde: PROPEX-RAG: Enhanced GraphRAG using Prompt-Driven Prompt Execution https://arxiv.org/abs/2511.01802 https://arxiv.org/pdf/2511.01802 https://arxiv.org/html/2511.01802
November 4, 2025 at 6:33 AM
Tejas Sarnaik, Manan Shah, Ravi Hegde: PROPEX-RAG: Enhanced GraphRAG using Prompt-Driven Prompt Execution https://arxiv.org/abs/2511.01802 https://arxiv.org/pdf/2511.01802 https://arxiv.org/html/2511.01802
今日のAI関連記事
「RAGとLLM、ベクトル検索…次はどう活用する？」～【python】今週の人気記事TOP5（2025/11/02） | Zennの「AI」のフィード
この記事は、RAGやLLM、ベクトル検索といったAI技術の応用をテーマにしたZennのPython人気記事トップ5を紹介する。
日本株分析アプリ開発、LLMの抽出処理高速化、Neo4jでのGraphRAG実装、動的な知識管理システム構築、言葉のベクトル化による3D可視化など、多様な技術活用事例を提示している。
「RAGとLLM、ベクトル検索…次はどう活用する？」～【python】今週の人気記事TOP5（2025/11/02） | Zennの「AI」のフィード
この記事は、RAGやLLM、ベクトル検索といったAI技術の応用をテーマにしたZennのPython人気記事トップ5を紹介する。
日本株分析アプリ開発、LLMの抽出処理高速化、Neo4jでのGraphRAG実装、動的な知識管理システム構築、言葉のベクトル化による3D可視化など、多様な技術活用事例を提示している。
「RAGとLLM、ベクトル検索…次はどう活用する？」～【python】今週の人気記事TOP5（2025/11/02）
!この記事はZennからpythonのLike数が多い記事のリストを自動的取得し、AIを利用して要約し、自動更新されています。 【2025/11/2】「RAGとLLM、ベクトル検索…次はどう活用する？」今週の人気記事TOP5（2025/11/02） 日本株3700社以上を分析。yfinance x「わが投資術」株式スクリーニングアプリを作った話（バイブコーディング）フルスタックエンジニアのテストく
zenn.dev
November 3, 2025 at 7:16 AM
今日のAI関連記事
「RAGとLLM、ベクトル検索…次はどう活用する？」～【python】今週の人気記事TOP5（2025/11/02） | Zennの「AI」のフィード
この記事は、RAGやLLM、ベクトル検索といったAI技術の応用をテーマにしたZennのPython人気記事トップ5を紹介する。
日本株分析アプリ開発、LLMの抽出処理高速化、Neo4jでのGraphRAG実装、動的な知識管理システム構築、言葉のベクトル化による3D可視化など、多様な技術活用事例を提示している。
「RAGとLLM、ベクトル検索…次はどう活用する？」～【python】今週の人気記事TOP5（2025/11/02） | Zennの「AI」のフィード
この記事は、RAGやLLM、ベクトル検索といったAI技術の応用をテーマにしたZennのPython人気記事トップ5を紹介する。
日本株分析アプリ開発、LLMの抽出処理高速化、Neo4jでのGraphRAG実装、動的な知識管理システム構築、言葉のベクトル化による3D可視化など、多様な技術活用事例を提示している。
https://zenn.dev/timelab/articles/23c0705465e0b4
この記事では、Neo4jを使用したGraphRAGの入門について解説されています。
GraphRAGは、グラフデータベースを活用したRAG（Retrieval-Augmented Generation）の手法です。
Neo4jとLangChainを連携させる方法が具体的に説明されています。
この記事では、Neo4jを使用したGraphRAGの入門について解説されています。
GraphRAGは、グラフデータベースを活用したRAG（Retrieval-Augmented Generation）の手法です。
Neo4jとLangChainを連携させる方法が具体的に説明されています。
Neo4jで始めるGraphRAG入門
zenn.dev
November 1, 2025 at 7:39 AM
https://zenn.dev/timelab/articles/23c0705465e0b4
この記事では、Neo4jを使用したGraphRAGの入門について解説されています。
GraphRAGは、グラフデータベースを活用したRAG（Retrieval-Augmented Generation）の手法です。
Neo4jとLangChainを連携させる方法が具体的に説明されています。
この記事では、Neo4jを使用したGraphRAGの入門について解説されています。
GraphRAGは、グラフデータベースを活用したRAG（Retrieval-Augmented Generation）の手法です。
Neo4jとLangChainを連携させる方法が具体的に説明されています。
Neo4jで始めるGraphRAG入門
https://zenn.dev/timelab/articles/23c0705465e0b4
https://zenn.dev/timelab/articles/23c0705465e0b4
Neo4jで始めるGraphRAG入門
zenn.dev
November 1, 2025 at 12:14 AM
Neo4jで始めるGraphRAG入門
https://zenn.dev/timelab/articles/23c0705465e0b4
https://zenn.dev/timelab/articles/23c0705465e0b4
"Build GraphRAG Applications with Amazon Neptune and Amazon Bedrock" by Kevin
#amazon-neptune #neptune #generative-ai
#amazon-neptune #neptune #generative-ai
Build GraphRAG Applications with Amazon Neptune and Amazon Bedrock
GraphRAG revolutionizes AI knowledge retrieval by combining Amazon Neptune's graph technology with Bedrock's foundation models, delivering responses that understand not just what information is similar, but how it's all connected - making AI answers smarter and more accurate.
community.aws
October 30, 2025 at 11:30 AM
"Build GraphRAG Applications with Amazon Neptune and Amazon Bedrock" by Kevin
#amazon-neptune #neptune #generative-ai
#amazon-neptune #neptune #generative-ai
Full steam ahead! A new #goingmeta episode is coming your way next Tuesday:
youtube.com/live/wFfld2v... #neo4j #knowledgegraph #graphrag
youtube.com/live/wFfld2v... #neo4j #knowledgegraph #graphrag
Going Meta S03E02 – a Series on Semantics, Knowledge Graphs and All Things AI
Season 03 Episode 2 of Going Meta – a Series on Semantics, Knowledge Graphs and All Things AI
Topic:
Jesús Barrasa: https://twitter.com/BarrasaDV
Repository: https://github.com/jbarrasa/goingmeta
K...
youtube.com
October 30, 2025 at 9:15 AM
Full steam ahead! A new #goingmeta episode is coming your way next Tuesday:
youtube.com/live/wFfld2v... #neo4j #knowledgegraph #graphrag
youtube.com/live/wFfld2v... #neo4j #knowledgegraph #graphrag
This implementation by Tomaz reverse-engineers @Microsoft's GraphRAG approach, adapted specifically for LlamaIndex workflows and neo4j. The system orchestrates entity extraction, community summarization, and dynamic knowledge graph traversal through several workflow steps.
October 29, 2025 at 6:06 PM
This implementation by Tomaz reverse-engineers @Microsoft's GraphRAG approach, adapted specifically for LlamaIndex workflows and neo4j. The system orchestrates entity extraction, community summarization, and dynamic knowledge graph traversal through several workflow steps.
Learn how to implement DRIFT Search with @neo4j and agent Workflows - a hybrid approach that combines global and local search for more accurate GraphRAG responses.
October 29, 2025 at 6:06 PM
Learn how to implement DRIFT Search with @neo4j and agent Workflows - a hybrid approach that combines global and local search for more accurate GraphRAG responses.
Graph-Guided Concept Selection for Efficient Retrieval-Augmented Generation
Huawei combines core chunk selection with an LLM-independent concept graph to reduce GraphRAG construction costs by 80% while improving retrieval effectiveness.
📝 arxiv.org/abs/2510.24120
Huawei combines core chunk selection with an LLM-independent concept graph to reduce GraphRAG construction costs by 80% while improving retrieval effectiveness.
📝 arxiv.org/abs/2510.24120
Graph-Guided Concept Selection for Efficient Retrieval-Augmented Generation
Graph-based RAG constructs a knowledge graph (KG) from text chunks to enhance retrieval in Large Language Model (LLM)-based question answering. It is especially beneficial in domains such as biomedici...
arxiv.org
October 29, 2025 at 4:03 AM
Graph-Guided Concept Selection for Efficient Retrieval-Augmented Generation
Huawei combines core chunk selection with an LLM-independent concept graph to reduce GraphRAG construction costs by 80% while improving retrieval effectiveness.
📝 arxiv.org/abs/2510.24120
Huawei combines core chunk selection with an LLM-independent concept graph to reduce GraphRAG construction costs by 80% while improving retrieval effectiveness.
📝 arxiv.org/abs/2510.24120
LLM이 지식 그래프 구축을 어떻게 혁신하고 있는지 살펴봅니다. 온톨로지 자동화, 스키마 프리 추출, 동적 메모리 시스템 등 최신 프레임워크(EDC, AutoSchemaKG, GraphRAG)를 통해 규칙 기반에서 생성 기반으로의 패러다임 전환을 소개합니다.
온톨로지 전문가가 사라진다: LLM이 바꾸는 지식 그래프의 미래
LLM이 지식 그래프 구축을 어떻게 혁신하고 있는지 살펴봅니다. 온톨로지 자동화, 스키마 프리 추출, 동적 메모리 시스템 등 최신 프레임워크(EDC, AutoSchemaKG, GraphRAG)를 통해 규칙 기반에서 생성 기반으로의 패러다임 전환을 소개합니다.
aisparkup.com
October 27, 2025 at 3:05 AM
LLM이 지식 그래프 구축을 어떻게 혁신하고 있는지 살펴봅니다. 온톨로지 자동화, 스키마 프리 추출, 동적 메모리 시스템 등 최신 프레임워크(EDC, AutoSchemaKG, GraphRAG)를 통해 규칙 기반에서 생성 기반으로의 패러다임 전환을 소개합니다.
Continuing to explore the seam between structure and statistics, this Fri 10AM we'll discuss “GraphRAG on Technical Documents - Impact of Knowledge Graph Schema” from the latest issue of Transactions on Graph Data and Knowledge.
drops.dagstuhl.de/entities/doc...
drops.dagstuhl.de/entities/doc...
GraphRAG on Technical Documents - Impact of Knowledge Graph Schema
drops.dagstuhl.de
October 22, 2025 at 3:43 PM
Continuing to explore the seam between structure and statistics, this Fri 10AM we'll discuss “GraphRAG on Technical Documents - Impact of Knowledge Graph Schema” from the latest issue of Transactions on Graph Data and Knowledge.
drops.dagstuhl.de/entities/doc...
drops.dagstuhl.de/entities/doc...
💡 RAG is not enough for building #LLM applications: hallucinations or long queries bog down performance are some of the results.
What do you need? Advanced RAG techniques.⚠️
https://bit.ly/4o0856t
#GraphRAG #Neo4j
What do you need? Advanced RAG techniques.⚠️
https://bit.ly/4o0856t
#GraphRAG #Neo4j
October 22, 2025 at 3:27 PM
💡 RAG is not enough for building #LLM applications: hallucinations or long queries bog down performance are some of the results.
What do you need? Advanced RAG techniques.⚠️
https://bit.ly/4o0856t
#GraphRAG #Neo4j
What do you need? Advanced RAG techniques.⚠️
https://bit.ly/4o0856t
#GraphRAG #Neo4j
In the battle of retrieval-augmented generation, who will win: GraphRAG or vector RAG?
We broke down both side by side to see which one’s better and help you make the decision easier.
Dive into the ultimate showdown 👉: www.meilisearch.com/blog/graph-r...
We broke down both side by side to see which one’s better and help you make the decision easier.
Dive into the ultimate showdown 👉: www.meilisearch.com/blog/graph-r...
GraphRAG vs. Vector RAG: Side-by-side comparison guide
A practical guide comparing GraphRAG and Vector RAG – how they work, key differences, pros/cons, top tools, and when to combine them for better answers.
www.meilisearch.com
October 22, 2025 at 3:04 PM
In the battle of retrieval-augmented generation, who will win: GraphRAG or vector RAG?
We broke down both side by side to see which one’s better and help you make the decision easier.
Dive into the ultimate showdown 👉: www.meilisearch.com/blog/graph-r...
We broke down both side by side to see which one’s better and help you make the decision easier.
Dive into the ultimate showdown 👉: www.meilisearch.com/blog/graph-r...
AI知識圖譜 GraphRAG 是怎麼回事？
[李東昇的說明]14分鐘，讓你了解 知識圖譜 如何跟 大語言模型 結合運用 ( 即 GraphRAG )
而 NER 就是 【命名實體辨識】
還有 關鍵是能 【舉一反三 觸類旁通】
youtu.be/WoU7XxDafbA?...
[李東昇的說明]14分鐘，讓你了解 知識圖譜 如何跟 大語言模型 結合運用 ( 即 GraphRAG )
而 NER 就是 【命名實體辨識】
還有 關鍵是能 【舉一反三 觸類旁通】
youtu.be/WoU7XxDafbA?...
AI知识图谱 GraphRAG 是怎么回事？
YouTube video by 程序员老王
youtu.be
October 20, 2025 at 1:45 PM
AI知識圖譜 GraphRAG 是怎麼回事？
[李東昇的說明]14分鐘，讓你了解 知識圖譜 如何跟 大語言模型 結合運用 ( 即 GraphRAG )
而 NER 就是 【命名實體辨識】
還有 關鍵是能 【舉一反三 觸類旁通】
youtu.be/WoU7XxDafbA?...
[李東昇的說明]14分鐘，讓你了解 知識圖譜 如何跟 大語言模型 結合運用 ( 即 GraphRAG )
而 NER 就是 【命名實體辨識】
還有 關鍵是能 【舉一反三 觸類旁通】
youtu.be/WoU7XxDafbA?...
🧠 @arcadedb.bsky.social now runs fully inside #Python with native bindings — no server, Docker, or network calls! Multi-model: graph, doc, vector, time-series. #Gremlin, #Cypher, #GraphQL, #SQL. Needs #JRE. 3 flavors.
🔗 Link in first 💬⤵️
Repost 🔁 #AI #RAG #GraphDatabase #JPype #GraphRAG #GraphDB
🔗 Link in first 💬⤵️
Repost 🔁 #AI #RAG #GraphDatabase #JPype #GraphRAG #GraphDB
October 18, 2025 at 8:53 PM
GraphRAG is tight. RAG is not. Google’s AI is a coward that won’t answer questions straight. ChatGPT is wildly inaccurate by default.
My cheap GraphRAG system on the other hand gives me correct and precise answers.
Bad AI is a choice that big companies make.
My cheap GraphRAG system on the other hand gives me correct and precise answers.
Bad AI is a choice that big companies make.
October 17, 2025 at 2:33 PM
GraphRAG is tight. RAG is not. Google’s AI is a coward that won’t answer questions straight. ChatGPT is wildly inaccurate by default.
My cheap GraphRAG system on the other hand gives me correct and precise answers.
Bad AI is a choice that big companies make.
My cheap GraphRAG system on the other hand gives me correct and precise answers.
Bad AI is a choice that big companies make.
Transactions on Graph Data and Knowledge, Volume 3, Issue 2 is now published! This issue includes three articles covering knowledge graph embeddings, formal concept analysis, and GraphRAG. Articles are published under Diamond OA, without fees for authors/readers:
drops.dagstuhl.de/entities/iss...
drops.dagstuhl.de/entities/iss...
TGDK, Volume 3, Issue 2
drops.dagstuhl.de
October 16, 2025 at 5:17 PM
Transactions on Graph Data and Knowledge, Volume 3, Issue 2 is now published! This issue includes three articles covering knowledge graph embeddings, formal concept analysis, and GraphRAG. Articles are published under Diamond OA, without fees for authors/readers:
drops.dagstuhl.de/entities/iss...
drops.dagstuhl.de/entities/iss...
LinearRAG: Linear Graph Retrieval Augmented Generation on Large-Scale Corpora
Constructs relation-free hierarchical graphs using lightweight entity extraction, reducing indexing time by over 77% while outperforming existing GraphRAG methods.
📝 arxiv.org/abs/2510.10114
Constructs relation-free hierarchical graphs using lightweight entity extraction, reducing indexing time by over 77% while outperforming existing GraphRAG methods.
📝 arxiv.org/abs/2510.10114
LinearRAG: Linear Graph Retrieval Augmented Generation on Large-scale Corpora
Retrieval-Augmented Generation (RAG) is widely used to mitigate hallucinations of Large Language Models (LLMs) by leveraging external knowledge. While effective for simple queries, traditional RAG sys...
arxiv.org
October 14, 2025 at 5:58 AM
LinearRAG: Linear Graph Retrieval Augmented Generation on Large-Scale Corpora
Constructs relation-free hierarchical graphs using lightweight entity extraction, reducing indexing time by over 77% while outperforming existing GraphRAG methods.
📝 arxiv.org/abs/2510.10114
Constructs relation-free hierarchical graphs using lightweight entity extraction, reducing indexing time by over 77% while outperforming existing GraphRAG methods.
📝 arxiv.org/abs/2510.10114