GraphRAG: Knowledge Graphs for AI Applications with Kirk Marple - 681

Today we're joined by Kirk Marple, CEO and founder of Graphlit, to explore the emerging paradigm of "GraphRAG," or Graph Retrieval Augmented Generation. In our conversation, Kirk digs into the GraphRAG architecture and how Graphlit uses it to offer a multi-stage workflow for ingesting, processing, retrieving, and generating content using LLMs (like GPT-4) and other Generative AI tech. He shares how the system performs entity extraction to build a knowledge graph and how graph, vector, and object storage are integrated in the system. We dive into how the system uses “prompt compilation” to improve the results it gets from Large Language Models during generation. We conclude by discussing several use cases the approach supports, as well as future agent-based applications it enables.

🗣️ CONNECT WITH US!
===============================

📖 CHAPTERS
===============================
00:00 - Introduction
03:31 - GraphRag
08:58 - Graphlit workflow
13:40 - Vector databases and graph databases
20:07 - Graphlit data retrieval
25:49 - Evaluation
29:37 - Different models
33:49 - Graphlit approach
36:20 - RAG outside of chatbots
44:29 - Graphlit details
46:05 - Conclusion

🔗 LINKS & RESOURCES
===============================

[jeffsteyn]

Xml and json. Nice think about json and xml once its outputted you can immediately use it in your code. You can also give 5 or more examples. The model becomes much more consistent. And in a specific task you can get similar performance to bigger more expensive models

[tuba-inxs]

Is Knowledge Graph, as it is used in this video an RDF graph?

[fintech]

agents are not very far though, its 'gradually, then suddenly'

[Tom_Goodwin]

Heya Sam just to let you know the Apple Podcasts version is only 4 mins long! That’s what sent me here aha