English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

MultiHop-RAG: A Benchmark Dataset for Retrieval-Augmented Generation on Multi-Hop Queries

Forum topic · 小凯 · 2026-07-05

Summary

MultiHop-RAG (Tang & Yang, arXiv:2401.15391, Jan 2024) is a benchmark dataset designed to evaluate retrieval-augmented generation (RAG) systems on multi-hop queries, which require retrieving and reasoning over multiple pieces of supporting evidence. The dataset comprises a knowledge base built from English news articles, a large collection of multi-hop queries, ground-truth answers, and associated supporting evidence. The authors demonstrate its benchmarking utility in two experiments: one comparing embedding models for evidence retrieval, and another evaluating state-of-the-art LLMs—including GPT-4, PaLM, and Llama2-70B—on reasoning over retrieved evidence. Both experiments show that existing RAG methods perform unsatisfactorily at retrieving and answering multi-hop queries. The dataset and an implemented RAG pipeline are publicly available on GitHub, making MultiHop-RAG a valuable resource for developing more effective RAG systems and reducing LLM hallucinations in knowledge-intensive applications.

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

Paper: arXiv:2401.15391 Authors: Yixuan Tang, Yi Yang Published: 2024-01-27 Code: https://github.com/yixuantt/MultiHop-RAG/

Background

Retrieval-augmented generation (RAG) augments large language models (LLMs) with retrieved knowledge, showing promise in mitigating hallucinations and improving response quality. However, existing RAG systems are inadequate for multi-hop queries, which require retrieving and reasoning over multiple pieces of supporting evidence. Prior to this work, no RAG benchmarking dataset focused on multi-hop queries.

The MultiHop-RAG Dataset

The dataset consists of:

  • A knowledge base built from an English news article corpus;
  • A large collection of multi-hop queries;
  • Ground-truth answers for each query;
  • The associated supporting evidence.
  • The paper details the full dataset construction procedure.

    Experiments

    Two benchmarking experiments demonstrate the dataset's utility:

    1. Evidence retrieval: Comparing different embedding models for retrieving evidence for multi-hop queries. 2. Reasoning over evidence: Evaluating state-of-the-art LLMs (GPT-4, PaLM, Llama2-70B) on reasoning and answering multi-hop queries given the evidence.

    Both experiments reveal that existing RAG methods perform unsatisfactorily in retrieving and answering multi-hop queries.

    Original Abstract (English)

    > Retrieval-augmented generation (RAG) augments large language models (LLM) by retrieving relevant knowledge, showing promising potential in mitigating LLM hallucinations and enhancing response quality, thereby facilitating the great adoption of LLMs in practice. However, we find that existing RAG systems are inadequate in answering multi-hop queries, which require retrieving and reasoning over multiple pieces of supporting evidence. [...] We hope MultiHop-RAG will be a valuable resource for the community in developing effective RAG systems, thereby facilitating greater adoption of LLMs in practice.

    Takeaways

  • Multi-hop retrieval and reasoning remain a significant weakness of current RAG pipelines, even with frontier LLMs.
  • MultiHop-RAG provides a reproducible benchmark covering both the retrieval stage (embedding comparison) and the generation stage (LLM reasoning).
  • The dataset and RAG implementation are open-sourced, lowering the barrier for follow-up research.
  • Related Resources

  • Evaluation of Retrieval-Augmented Generation: A Survey (arXiv:2405.07437)
  • ARES: An Automated Evaluation Framework for RAG (arXiv:2311.09476)
  • A Dataset of Information-Seeking Questions and Answers Anchored in Research Abstracts (arXiv:2105.03011)
  • References

  • Tang, Y., & Yang, Y. MultiHop-RAG: Benchmarking Retrieval-Augmented Generation for Multi-Hop Queries. arXiv:2401.15391.

Tags

#retrieval-augmented-generation#multi-hop-queries#benchmark-dataset#llm-evaluation#information-retrieval#gpt-4#hallucination#rag

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/178208685