Completed
[] Faithfulness measures the consistency of LLM-generated answers
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Integrating Knowledge Graphs and Vector RAG for Efficient Information Extraction - Reading Group Session
Automatically move to the next video in the Classroom when playback concludes
- 1 [] Hybrid rank combines knowledge graphs, vector retrieval
- 2 [] Hybrid drag improves accuracy using knowledge graphs
- 3 [] Text preprocessing and graph algorithms improve LLM responses
- 4 [] Faithfulness measures the consistency of LLM-generated answers
- 5 [] Faithfulness metric should normalize statements' word counts
- 6 [] Analyzed Nifty 50 earning calls for hedge fund
- 7 [] Construct a knowledge graph and chunk text data
- 8 [] Example of earnings extraction illustrating graph advantages
- 9 [] Traversing graph to find relevant entities' relationships
- 10 [] Highly connected graph; combine vectors, graphs arbitrarily
- 11 [] Graph neural networks improve re-ranking and retrieval
- 12 [] Comparison methods are unfair due to unequal context size
- 13 [] Making larger language models traverse graph structures
- 14 [] Consider separating entity attributes from relationships
- 15 [] Join discussions on Slack, suggest topics, participate
- 16 [] Wrap up
