Multi-lingual Discourse Segmentation and Connective Identification: MELODI at Disrpt2021 - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2021

Multi-lingual Discourse Segmentation and Connective Identification: MELODI at Disrpt2021

Résumé

We present an approach for discourse segmentation and discourse connective identification, both at the sentence and document level, within the DISRPT 2021 shared task, a multilingual and multi-formalism evaluation campaign. 1 Building on the most successful architecture from the 2019 similar shared task, we leverage datasets in the same or similar languages to augment training data and improve on the best systems from the previous campaign on 3 out of 4 subtasks, with a mean improvement on all 16 datasets of 0.85%. Within the Disrpt 21 campaign the system ranks 3rd overall, very close to the 2nd system, but with a significant gap with respect to the best system, which uses a rich set of additional features. The system is nonetheless the best on languages that benefited from crosslingual training on sentence internal segmentation (German and Spanish).
Fichier principal
Vignette du fichier
2021.disrpt-1.3.pdf (199.9 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03725591 , version 1 (17-07-2022)

Identifiants

Citer

Morteza Kamaladdini Ezzabady, Philippe Muller, Chloé Braud. Multi-lingual Discourse Segmentation and Connective Identification: MELODI at Disrpt2021. 2nd Shared Task on Discourse Relation Parsing and Treebanking (DISRPT 2021), Nov 2021, Punta Cana, Dominican Republic. pp.22-32, ⟨10.18653/v1/2021.disrpt-1.3⟩. ⟨hal-03725591⟩
38 Consultations
56 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More