Anaphora Resolution in Dialogue: System Description (CODI-CRAC 2022 Shared Task) - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2022

Anaphora Resolution in Dialogue: System Description (CODI-CRAC 2022 Shared Task)

Tatiana Anikina
  • Fonction : Auteur
  • PersonId : 1212174
Natalia Skachkova
  • Fonction : Auteur
  • PersonId : 1212175

Résumé

We describe three models submitted for the CODI-CRAC 2022 shared task. To perform identity anaphora resolution, we test several combinations of the incremental clustering approach based on the Workspace Coreference System (WCS) with other coreference models. The best result is achieved by adding the "cluster merging" version of the coref-hoi model, which brings up to 10.33% improvement 1 over vanilla WCS clustering. Discourse deixis resolution is implemented as multi-task learning: we combine the learning objective of corefhoi with anaphor type classification. We adapt the higher-order resolution model introduced in Joshi et al. (2019) for bridging resolution given gold mentions and anaphors.
Fichier principal
Vignette du fichier
2022.codi-crac.2.pdf (275.52 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03925147 , version 1 (05-01-2023)

Identifiants

Citer

Tatiana Anikina, Natalia Skachkova, Joseph Renner, Priyansh Trivedi. Anaphora Resolution in Dialogue: System Description (CODI-CRAC 2022 Shared Task). CODI-CRAC 2022 - 5th Workshop on Computational Models of Reference, Anaphora and Coreference, Oct 2022, Gyeongju, South Korea. ⟨hal-03925147⟩
31 Consultations
27 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More