DMRST: A Joint Framework for Document-Level Multilingual RST Discourse Segmentation and Parsing

Abstract

Text discourse parsing weighs importantly in understanding information flow and argumentative structure in natural language, making it beneficial for downstream tasks. While previous work significantly improves the performance of RST discourse parsing, they are not readily applicable to practical use cases: (1) EDU segmentation is not integrated into most existing tree parsing frameworks, thus it is not straightforward to apply such models on newly-coming data. (2) Most parsers cannot be used in multilingual scenarios, because they are developed only in English. (3) Parsers trained from single-domain treebanks do not generalize well on out-of-domain inputs. In this work, we propose a document-level multilingual RST discourse parsing framework, which conducts EDU segmentation and discourse tree parsing jointly. Moreover, we propose a cross-translation augmentation strategy to enable the framework to support multilingual parsing and improve its domain generality. Experimental results show that our model achieves state-of-the-art performance on document-level multilingual RST parsing in all sub-tasks.

Semi-supervised News Discourse Profiling with Contrastive Learning

“…News discourse profiling (Choubey et al, 2020) is a specialized task aimed at comprehensively analyzing the structural aspects of news articles and effectively categorizing each sentence based on its contextual depiction of news events. Therefore, this is a document-level task with sentence-level predictions (Li et al, 2022a), which has been proven useful in several downstream tasks, including text simplification (Zhang et al, 2022a), media bias analysis (Lei et al, 2022), event coreference resolution (Choubey et al, 2020), RST-style Discourse Parsing (Li and Huang, 2023) and temporal dependency graph building (Choubey and .…”

Discourse-aware text classification for argument mining

“…Interpreting discourse accurately may require sophisticated skills, such as reasoning over general knowledge and assessing the subjective significance of particular statements. Endto-end discourse tree prediction recently achieved 50.1% F1 on the RST-DT corpus (Liu et al, 2021). Discourse parsing is also significantly affected by domain shift.…”