Logo Lanfrica
  • Accueil
  • Atlas
  • Analyses
  • Documentation
  • Sign in

© 2026 Lanfrica. Tous droits réservés. Tous les droits d'auteur des ressources affichées sur le site Web Lanfrica appartiennent aux détenteurs de droits d'auteur d'origine, sauf indication contraire explicite.

DeDisCo at the DISRPT 2025 Shared Task: A System for Discourse Relation Classification

Domaine:

natural language processing

Type de record:

papersoftware
Créateur:
Ju,Wu,PurZel
Hôte:avatar
This paper presents DeDisCo, Georgetown University's entry in the DISRPT 2025 shared task on discourse relation classification. We test two approaches, using an mt5-based encoder and a decoder based approach using the openly available Qwen model. We also experiment on training with augmented dataset for low-resource languages using matched data translated automatically from English, as well as using some additional linguistic features inspired by entries in previous editions of the Shared Task. Our system achieves a macro-accuracy score of 71.28, and we provide some interpretation and error analysis for our results. System submission for the DISRPT 2025 - Shared Task on Discourse Relation Parsing and Treebanking In conjunction with CODI-CRAC & EMNLP 2025. 1st place in Task 3: relation classification

Visit

arxiv.org

Tags

Computation and Language

Similaires

Implicit Discourse Relation Classification For Nigerian PidginGMU Systems for the IWSLT 2025 Low-Resource Speech Translation Shared TaskNADI 2025: The First Multidialectal Arabic Speech Processing Shared TaskLIA and ELYADATA systems for the IWSLT 2025 low-resource speech translation shared taskEfficient Multi-Task Arabic Question Classification with a Shared MARBERT EncoderThe Shared Task on Gender Rewriting

Implicit Discourse Relation Classification For Nigerian Pidgin

Despite attempts to make Large Language Models multi-lingual, many of the world's languages are stil

GMU Systems for the IWSLT 2025 Low-Resource Speech Translation Shared Task

This paper describes the GMU systems for the IWSLT 2025 low-resource speech translation shared task.

NADI 2025: The First Multidialectal Arabic Speech Processing Shared Task

We present the findings of the sixth Nuanced Arabic Dialect Identification (NADI 2025) Shared Task,

LIA and ELYADATA systems for the IWSLT 2025 low-resource speech translation shared task

International audience

In this paper, we present the approach and system setu

Efficient Multi-Task Arabic Question Classification with a Shared MARBERT Encoder

No description provided. If you use this software, please cite the associated paper.

The Shared Task on Gender Rewriting

In this paper, we present the results and findings of the Shared Task on Gender Rewriting, which was