CLAPNQ: Cohesive Long-form Answers from Passages in Natural Questions for RAG systems

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Rosenthal, Sara, Sil, Avirup, Florian, Radu, Roukos, Salim
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915076486725632
author Rosenthal, Sara
Sil, Avirup
Florian, Radu
Roukos, Salim
author_facet Rosenthal, Sara
Sil, Avirup
Florian, Radu
Roukos, Salim
contents Retrieval Augmented Generation (RAG) has become a popular application for large language models. It is preferable that successful RAG systems provide accurate answers that are supported by being grounded in a passage without any hallucinations. While considerable work is required for building a full RAG pipeline, being able to benchmark performance is also necessary. We present ClapNQ, a benchmark Long-form Question Answering dataset for the full RAG pipeline. ClapNQ includes long answers with grounded gold passages from Natural Questions (NQ) and a corpus to perform either retrieval, generation, or the full RAG pipeline. The ClapNQ answers are concise, 3x smaller than the full passage, and cohesive, meaning that the answer is composed fluently, often by integrating multiple pieces of the passage that are not contiguous. RAG models must adapt to these properties to be successful at ClapNQ. We present baseline experiments and analysis for ClapNQ that highlight areas where there is still significant room for improvement in grounded RAG. CLAPNQ is publicly available at https://github.com/primeqa/clapnq
format Preprint
id arxiv_https___arxiv_org_abs_2404_02103
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle CLAPNQ: Cohesive Long-form Answers from Passages in Natural Questions for RAG systems
Rosenthal, Sara
Sil, Avirup
Florian, Radu
Roukos, Salim
Computation and Language
Retrieval Augmented Generation (RAG) has become a popular application for large language models. It is preferable that successful RAG systems provide accurate answers that are supported by being grounded in a passage without any hallucinations. While considerable work is required for building a full RAG pipeline, being able to benchmark performance is also necessary. We present ClapNQ, a benchmark Long-form Question Answering dataset for the full RAG pipeline. ClapNQ includes long answers with grounded gold passages from Natural Questions (NQ) and a corpus to perform either retrieval, generation, or the full RAG pipeline. The ClapNQ answers are concise, 3x smaller than the full passage, and cohesive, meaning that the answer is composed fluently, often by integrating multiple pieces of the passage that are not contiguous. RAG models must adapt to these properties to be successful at ClapNQ. We present baseline experiments and analysis for ClapNQ that highlight areas where there is still significant room for improvement in grounded RAG. CLAPNQ is publicly available at https://github.com/primeqa/clapnq
title CLAPNQ: Cohesive Long-form Answers from Passages in Natural Questions for RAG systems
topic Computation and Language
url https://arxiv.org/abs/2404.02103