TIGER-AI-Lab/LongRAG
Official repo for "LongRAG: Enhancing Retrieval-Augmented Generation with Long-context LLMs".
Implements a two-stage architecture pairing long retrieval units (4K tokens, 30× longer than traditional approaches) with long-context LLM readers to rebalance retriever-reader workload asymmetry. Leverages Tevatron for dense retrieval with BGE embeddings and integrates with frontier LLMs (GPT-4o, Gemini-1.5-Pro, Claude-3-Opus) for reading comprehension. Provides preprocessed corpora and evaluation scripts for NQ and HotpotQA datasets via Hugging Face.
245 stars. No commits in the last 6 months.
Stars
245
Forks
19
Language
Python
License
MIT
Category
Last pushed
Aug 25, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/rag/TIGER-AI-Lab/LongRAG"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
VectorInstitute/fed-rag
A framework for fine-tuning retrieval-augmented generation (RAG) systems.
ictnlp/FlexRAG
FlexRAG: A RAG Framework for Information Retrieval and Generation.
gomate-community/TrustRAG
TrustRAG：The RAG Framework within Reliable input,Trusted output
NirDiamant/RAG_Techniques
This repository showcases various advanced techniques for Retrieval-Augmented Generation (RAG)...
Andrew-Jang/RAGHub
A community-driven collection of RAG (Retrieval-Augmented Generation) frameworks, projects, and...