Granuscore: A Reference-Free Measure of Granularity for Text Analysis and Question Answering

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Ellinger, Lukas, Fichtl, Alexander, Anschütz, Miriam, Groh, Georg
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918523734851584
author Ellinger, Lukas
Fichtl, Alexander
Anschütz, Miriam
Groh, Georg
author_facet Ellinger, Lukas
Fichtl, Alexander
Anschütz, Miriam
Groh, Georg
contents Natural language conveys information at varying levels of granularity, from fine-grained references to broad descriptions. While granularity is fundamental to human communication, existing measures mostly capture surface detail or sentence specificity. We introduce Granuscore, a reference-free measure of granularity that leverages structural properties of a hierarchical embedding space. Granuscore reliably recovers hierarchical orderings on the Granola-EQ dataset and captures expected differences in granularity across discourse contexts. Across domains, we further show that Granuscore explains non-linear variation in sentence specificity beyond sentence length. Finally, we apply Granuscore to four question-answering benchmarks and analyze how granularity differs for questions, gold answers, and model outputs across response outcomes. The analysis reveals consistent differences in model behavior and provides a principled lens for characterizing the difficulty of QA datasets. Together, the results position Granuscore as a scalable, broadly applicable tool for analyzing granularity in text.
format Preprint
id arxiv_https___arxiv_org_abs_2605_26620
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Granuscore: A Reference-Free Measure of Granularity for Text Analysis and Question Answering
Ellinger, Lukas
Fichtl, Alexander
Anschütz, Miriam
Groh, Georg
Computation and Language
Human-Computer Interaction
Natural language conveys information at varying levels of granularity, from fine-grained references to broad descriptions. While granularity is fundamental to human communication, existing measures mostly capture surface detail or sentence specificity. We introduce Granuscore, a reference-free measure of granularity that leverages structural properties of a hierarchical embedding space. Granuscore reliably recovers hierarchical orderings on the Granola-EQ dataset and captures expected differences in granularity across discourse contexts. Across domains, we further show that Granuscore explains non-linear variation in sentence specificity beyond sentence length. Finally, we apply Granuscore to four question-answering benchmarks and analyze how granularity differs for questions, gold answers, and model outputs across response outcomes. The analysis reveals consistent differences in model behavior and provides a principled lens for characterizing the difficulty of QA datasets. Together, the results position Granuscore as a scalable, broadly applicable tool for analyzing granularity in text.
title Granuscore: A Reference-Free Measure of Granularity for Text Analysis and Question Answering
topic Computation and Language
Human-Computer Interaction
url https://arxiv.org/abs/2605.26620