BaranziniLab/KG_RAG
A knowledge graph-based retrieval-augmented generation system that enhances LLMs for domain-specific question answering.

Velocity · 7d
+1.0
★ / day
Trend
→steady
star history
The system retrieves structured knowledge from knowledge graphs to augment LLM prompts, improving accuracy on knowledge-intensive tasks. It integrates with multiple LLM providers (GPT-3.5/4, Llama) and can be configured for domain-specific applications. Includes a benchmark dataset (BiomixQA) for evaluation.