Year: 2025
Type: conference-paper
Abstract: Question Under Discussion (QUD) is a discourse framework that has attracted growing interest in NLP in recent years.Among existing QUD models, the QUD tree approach (Riester, 2019) focuses on reconstructing QUDs and their hierarchical relationships, using a single tree to represent discourse structure.Prior implementation shows moderate inter-annotator agreement, highlighting the challenging nature of this task.In this paper, we propose a new QUD model for annotating hierarchical discourse structure.Our annotation achieves high inter-annotator agreement: 81.45% for short files and 79.53% for long files of Wall Street Journal articles.We show preliminary results on using GPT-4 for automatic annotation, which suggests that one of the best-performing LLMs still struggles with capturing hierarchical discourse structure.Moreover, we compare the annotations with RST annotations.Lastly, we present an approach for integrating hierarchical and local discourse relation annotations with the proposed model. * Equal contribution.another QUD, and multiple QUDs can jointly contribute to resolving a higher-level QUD 1 .Discourse structure can thus be understood through the relationships between these QUDs.An example is shown in Figure 1.With QUDs reconstructed and added to discourse, the texts remain coherent, and the information conveyed is largely unchanged.Moreover, the relationship between sentences is mirrored by the relationship between QUDs.
Authors: Yingxue Fu, Mark-Jan Nederhof, Anais Ollagnier
Institutions: Centre National de la Recherche Scientifique, Institut national de recherche en sciences et technologies du numérique, University of St Andrews, Université Côte d'Azur, Laboratoire d'Informatique, Signaux et Systèmes de Sophia Antipolis, MARIANNE: Modèles et données pour l'argumentation computationnelle en langage naturel
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W4415348387
FWCI:
Cites:
Cited by:
Related to: