Measuring Retrieval Complexity in Question Answering Systems

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Gabburo, Matteo, Jedema, Nicolaas Paul, Garg, Siddhant, Ribeiro, Leonardo F. R., Moschitti, Alessandro
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866914827039932416
author Gabburo, Matteo
Jedema, Nicolaas Paul
Garg, Siddhant
Ribeiro, Leonardo F. R.
Moschitti, Alessandro
author_facet Gabburo, Matteo
Jedema, Nicolaas Paul
Garg, Siddhant
Ribeiro, Leonardo F. R.
Moschitti, Alessandro
contents In this paper, we investigate which questions are challenging for retrieval-based Question Answering (QA). We (i) propose retrieval complexity (RC), a novel metric conditioned on the completeness of retrieved documents, which measures the difficulty of answering questions, and (ii) propose an unsupervised pipeline to measure RC given an arbitrary retrieval system. Our proposed pipeline measures RC more accurately than alternative estimators, including LLMs, on six challenging QA benchmarks. Further investigation reveals that RC scores strongly correlate with both QA performance and expert judgment across five of the six studied benchmarks, indicating that RC is an effective measure of question difficulty. Subsequent categorization of high-RC questions shows that they span a broad set of question shapes, including multi-hop, compositional, and temporal QA, indicating that RC scores can categorize a new subset of complex questions. Our system can also have a major impact on retrieval-based systems by helping to identify more challenging questions on existing datasets.
format Preprint
id arxiv_https___arxiv_org_abs_2406_03592
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Measuring Retrieval Complexity in Question Answering Systems
Gabburo, Matteo
Jedema, Nicolaas Paul
Garg, Siddhant
Ribeiro, Leonardo F. R.
Moschitti, Alessandro
Computation and Language
Artificial Intelligence
In this paper, we investigate which questions are challenging for retrieval-based Question Answering (QA). We (i) propose retrieval complexity (RC), a novel metric conditioned on the completeness of retrieved documents, which measures the difficulty of answering questions, and (ii) propose an unsupervised pipeline to measure RC given an arbitrary retrieval system. Our proposed pipeline measures RC more accurately than alternative estimators, including LLMs, on six challenging QA benchmarks. Further investigation reveals that RC scores strongly correlate with both QA performance and expert judgment across five of the six studied benchmarks, indicating that RC is an effective measure of question difficulty. Subsequent categorization of high-RC questions shows that they span a broad set of question shapes, including multi-hop, compositional, and temporal QA, indicating that RC scores can categorize a new subset of complex questions. Our system can also have a major impact on retrieval-based systems by helping to identify more challenging questions on existing datasets.
title Measuring Retrieval Complexity in Question Answering Systems
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2406.03592