LiteraryQA: Towards Effective Evaluation of Long-document Narrative QA

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Bonomo, Tommaso, Gioffré, Luca, Navigli, Roberto
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914094964015104
author Bonomo, Tommaso
Gioffré, Luca
Navigli, Roberto
author_facet Bonomo, Tommaso
Gioffré, Luca
Navigli, Roberto
contents Question Answering (QA) on narrative text poses a unique challenge to current systems, requiring a deep understanding of long, complex documents. However, the reliability of NarrativeQA, the most widely used benchmark in this domain, is hindered by noisy documents and flawed QA pairs. In this work, we introduce LiteraryQA, a high-quality subset of NarrativeQA focused on literary works. Using a human- and LLM-validated pipeline, we identify and correct low-quality QA samples while removing extraneous text from source documents. We then carry out a meta-evaluation of automatic metrics to clarify how systems should be evaluated on LiteraryQA. This analysis reveals that all n-gram-based metrics have a low system-level correlation to human judgment, while LLM-as-a-Judge evaluations, even with small open-weight models, can strongly agree with the ranking identified by humans. Finally, we benchmark a set of long-context LLMs on LiteraryQA. We release our code and data at https://github.com/SapienzaNLP/LiteraryQA.
format Preprint
id arxiv_https___arxiv_org_abs_2510_13494
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle LiteraryQA: Towards Effective Evaluation of Long-document Narrative QA
Bonomo, Tommaso
Gioffré, Luca
Navigli, Roberto
Computation and Language
Artificial Intelligence
Question Answering (QA) on narrative text poses a unique challenge to current systems, requiring a deep understanding of long, complex documents. However, the reliability of NarrativeQA, the most widely used benchmark in this domain, is hindered by noisy documents and flawed QA pairs. In this work, we introduce LiteraryQA, a high-quality subset of NarrativeQA focused on literary works. Using a human- and LLM-validated pipeline, we identify and correct low-quality QA samples while removing extraneous text from source documents. We then carry out a meta-evaluation of automatic metrics to clarify how systems should be evaluated on LiteraryQA. This analysis reveals that all n-gram-based metrics have a low system-level correlation to human judgment, while LLM-as-a-Judge evaluations, even with small open-weight models, can strongly agree with the ranking identified by humans. Finally, we benchmark a set of long-context LLMs on LiteraryQA. We release our code and data at https://github.com/SapienzaNLP/LiteraryQA.
title LiteraryQA: Towards Effective Evaluation of Long-document Narrative QA
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2510.13494