EfficientRAG: Efficient Retriever for Multi-Hop Question Answering

Ziyuan Zhuang, Zhiyang Zhang, Sitao Cheng, Fangkai Yang, Jia Liu +5 more
8/8/2024
cs.CLcs.AI

No official code found. Best implementation: NIL-zhuang/EfficientRAG-official (69 stars, MIT, last commit 2025-03-04).

Abstract

Retrieval-augmented generation (RAG) methods encounter difficulties when addressing complex questions like multi-hop queries. While iterative retrieval methods improve performance by gathering additional information, current approaches often rely on multiple calls of large language models (LLMs). In this paper, we introduce EfficientRAG, an efficient retriever for multi-hop question answering. EfficientRAG iteratively generates new queries without the need for LLM calls at each iteration and filters out irrelevant information. Experimental results demonstrate that EfficientRAG surpasses existing RAG methods on three open-domain multi-hop question-answering datasets.

View on arXivView PDF

Code Implementations(1)

NIL-zhuang/EfficientRAG-officialCommunity100%
MIT

Code Repo for EfficientRAG: Efficient Retriever for Multi-Hop Question Answering

696Shell, PythonSep 11, 20241 years agoMIT

Similar Papers

HANRAG: Heuristic Accurate Noise-resistant Retrieval-Augmented Generation for Multi-hop Question Answering

92% match2025

Retrieve, Summarize, Plan: Advancing Multi-hop Question Answering with an Iterative Approach

87% match2024

MultiHop-RAG: Benchmarking Retrieval-Augmented Generation for Multi-Hop Queries

87% match1 repo4682024

Optimizing Multi-Hop Document Retrieval Through Intermediate Representations

87% match2025

Enhancing Document-Level Question Answering via Multi-Hop Retrieval-Augmented Generation with LLaMA 3

87% match2025

Cooperative Retrieval-Augmented Generation for Question Answering: Mutual Information Exchange and Ranking by Contrasting Layers

86% match2025

Cite this paper

@article{zhuang2024efficientrag,
  title  = {EfficientRAG: Efficient Retriever for Multi-Hop Question Answering},
  author = {Ziyuan Zhuang and Zhiyang Zhang and Sitao Cheng and Fangkai Yang and Jia Liu and Shujian Huang and Qingwei Lin and Saravan Rajmohan and Dongmei Zhang and Qi Zhang},
  year   = {2024},
  eprint = {2408.04259},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2408.04259v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2408.04259.svg)](https://codeofpaper.com/paper/2408.04259)

Discussion