Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Martínez-Murillo, Iván, Moreda, Paloma, Lloret, Elena
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2510.24179
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866912673886633984
author Martínez-Murillo, Iván
Moreda, Paloma
Lloret, Elena
author_facet Martínez-Murillo, Iván
Moreda, Paloma
Lloret, Elena
contents This paper explores the influence of external knowledge integration in Natural Language Generation (NLG), focusing on a commonsense generation task. We extend the CommonGen dataset by creating KITGI, a benchmark that pairs input concept sets with retrieved semantic relations from ConceptNet and includes manually annotated outputs. Using the T5-Large model, we compare sentence generation under two conditions: with full external knowledge and with filtered knowledge where highly relevant relations were deliberately removed. Our interpretability benchmark follows a three-stage method: (1) identifying and removing key knowledge, (2) regenerating sentences, and (3) manually assessing outputs for commonsense plausibility and concept coverage. Results show that sentences generated with full knowledge achieved 91\% correctness across both criteria, while filtering reduced performance drastically to 6\%. These findings demonstrate that relevant external knowledge is critical for maintaining both coherence and concept coverage in NLG. This work highlights the importance of designing interpretable, knowledge-enhanced NLG systems and calls for evaluation frameworks that capture the underlying reasoning beyond surface-level metrics.
format Preprint
id arxiv_https___arxiv_org_abs_2510_24179
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Exploring the Influence of Relevant Knowledge for Natural Language Generation Interpretability
Martínez-Murillo, Iván
Moreda, Paloma
Lloret, Elena
Computation and Language
This paper explores the influence of external knowledge integration in Natural Language Generation (NLG), focusing on a commonsense generation task. We extend the CommonGen dataset by creating KITGI, a benchmark that pairs input concept sets with retrieved semantic relations from ConceptNet and includes manually annotated outputs. Using the T5-Large model, we compare sentence generation under two conditions: with full external knowledge and with filtered knowledge where highly relevant relations were deliberately removed. Our interpretability benchmark follows a three-stage method: (1) identifying and removing key knowledge, (2) regenerating sentences, and (3) manually assessing outputs for commonsense plausibility and concept coverage. Results show that sentences generated with full knowledge achieved 91\% correctness across both criteria, while filtering reduced performance drastically to 6\%. These findings demonstrate that relevant external knowledge is critical for maintaining both coherence and concept coverage in NLG. This work highlights the importance of designing interpretable, knowledge-enhanced NLG systems and calls for evaluation frameworks that capture the underlying reasoning beyond surface-level metrics.
title Exploring the Influence of Relevant Knowledge for Natural Language Generation Interpretability
topic Computation and Language
url https://arxiv.org/abs/2510.24179