Graph RAG with pure vector search, achieving SOTA performance in multi-hop reasoning scenarios.

Highly Efficient Query Rewriter for Passage Retrieval in the realm of Retrieval-Augmented Generation (RAG)

AT-RAG (Adaptive Retrieval-Augmented Generation) is a novel RAG model developed to address the challenges of complex multi-hop queries

Implemented a question and answering model for multi-hop questions that requires logical inference or aggregation of information from various parts of the information text (like referring multiple wikis to answer a question)

Enhancing Retrieval-Augmented Generation with Document Link Structure for Multi-hop Web Question Answering

Meandering In Networks of Entities to Reach Verisimilar Answers

MAT: Multihop Annotation Tool

Modifications to Fusion in Decoder architecture to make more it efficient