LongRAG

简介

📃Paper | 🌐Website | 💻Github | 🛢️Datasets Overview In traditional RAG framework, the basic retrieval units are normally short. Such a design forces the retriever to search over a large corpus to find the "needle" unit. In contrast, the readers only need to extract answers from the short retrieved units. Such an imbalanced heavy retriever and light reader design can lead to sub-optimal performance. We propose a new framework LongRAG, consisting of a "long retriever" and a "long… See the full description on the dataset page: https://huggingface.co/datasets/TIGER-Lab/LongRAG.