Contributo in atti di convegno, 2023, ENG, 10.18653/v1/2023.findings-acl.680

Coherent or Not? Stressing a Neural Language Model for Discourse Coherence in Multiple Languages

Dominique Brunato; Felice Dell'Orletta; Irene Dini; Andrea Amelio Ravelli

Istituto di Linguistica Computazionale "Antonio Zampolli", Pisa; Istituto di Linguistica Computazionale "Antonio Zampolli", Pisa; Istituto di Linguistica Computazionale "Antonio Zampolli", Pisa/ University of Pisa; Istituto di Linguistica Computazionale ILC-CNR / University of Bologna

In this study, we investigate the capability of a Neural Language Model (NLM) to distinguish between coherent and incoherent text, where the latter has been artificially created to gradually undermine local coherence within text. While previous research on coherence assessment using NLMs has primarily focused on English, we extend our investigation to multiple languages. We employ a consistent evaluation framework to compare the performance of monolingual and multilingual models in both in-domain and out-domain settings. Additionally, we explore the model's performance in a cross-language scenario.

61st Annual Meeting of the Association for Computational Linguistics (ACL 2023), pp. 10690–10700, Toronto, Canada, 9-14/07/2023

Keywords

text coherence, neural language models, multilingual corpora

CNR authors

Ravelli Andrea Amelio, Dini Irene, Dell Orletta Felice, Brunato Dominique Pierina

CNR institutes

ILC – Istituto di linguistica computazionale "Antonio Zampolli"

ID: 491078

Year: 2023

Type: Contributo in atti di convegno

Creation: 2024-01-03 13:47:53.000

Last update: 2024-01-10 11:02:52.000

External IDs

CNR OAI-PMH: oai:it.cnr:prodotti:491078

DOI: 10.18653/v1/2023.findings-acl.680