Bridging Discourse Treebanks with a Unified Rhetorical Structure Parser

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
1. Verfasser: Chistova, Elena
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915538429542400
author Chistova, Elena
author_facet Chistova, Elena
contents We introduce UniRST, the first unified RST-style discourse parser capable of handling 18 treebanks in 11 languages without modifying their relation inventories. To overcome inventory incompatibilities, we propose and evaluate two training strategies: Multi-Head, which assigns separate relation classification layer per inventory, and Masked-Union, which enables shared parameter training through selective label masking. We first benchmark monotreebank parsing with a simple yet effective augmentation technique for low-resource settings. We then train a unified model and show that (1) the parameter efficient Masked-Union approach is also the strongest, and (2) UniRST outperforms 16 of 18 mono-treebank baselines, demonstrating the advantages of a single-model, multilingual end-to-end discourse parsing across diverse resources.
format Preprint
id arxiv_https___arxiv_org_abs_2510_06427
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Bridging Discourse Treebanks with a Unified Rhetorical Structure Parser
Chistova, Elena
Computation and Language
We introduce UniRST, the first unified RST-style discourse parser capable of handling 18 treebanks in 11 languages without modifying their relation inventories. To overcome inventory incompatibilities, we propose and evaluate two training strategies: Multi-Head, which assigns separate relation classification layer per inventory, and Masked-Union, which enables shared parameter training through selective label masking. We first benchmark monotreebank parsing with a simple yet effective augmentation technique for low-resource settings. We then train a unified model and show that (1) the parameter efficient Masked-Union approach is also the strongest, and (2) UniRST outperforms 16 of 18 mono-treebank baselines, demonstrating the advantages of a single-model, multilingual end-to-end discourse parsing across diverse resources.
title Bridging Discourse Treebanks with a Unified Rhetorical Structure Parser
topic Computation and Language
url https://arxiv.org/abs/2510.06427