Labeling Explicit Discourse Relations Using Pre-trained Language Models

Kurfali, Murathan, 1990- (author): Stockholms universitet,Institutionen för lingvistik

(creator_code:org_t)

2020-09-01
2020
English.
In: Text, Speech, and Dialogue. - Cham : Springer. - 9783030583231 - 9783030583224 ; , s. 79-86

Related links:: http://arxiv.org/pdf...; show more...; https://urn.kb.se/re...; https://doi.org/10.1...; show less...

Abstract Subject headings

Labeling explicit discourse relations is one of the most challenging sub-tasks of the shallow discourse parsing where the goal is to identify the discourse connectives and the boundaries of their arguments. The state-of-the-art models achieve slightly above 45% of F-score by using hand-crafted features. The current paper investigates the efficacy of the pre-trained language models in this task. We find that the pre-trained language models, when finetuned, are powerful enough to replace the linguistic features. We evaluate our model on PDTB 2.0 and report the state-of-the-art results in extraction of the full relation. This is the first time when a model outperforms the knowledge intensive models without employing any linguistic features.

About the subject

NATURAL SCIENCES: NATURAL SCIENCES; and Computer and Inf ...; and Language Technol ...

Kungliga biblioteket hanterar dina personuppgifter i enlighet med EU:s dataskyddsförordning (2018), GDPR. Läs mer om hur det funkar här.
Så här hanterar KB dina uppgifter vid användning av denna tjänst.