English

Neural RST-based Evaluation of Discourse Coherence

Computation and Language 2020-10-01 v1

Abstract

This paper evaluates the utility of Rhetorical Structure Theory (RST) trees and relations in discourse coherence evaluation. We show that incorporating silver-standard RST features can increase accuracy when classifying coherence. We demonstrate this through our tree-recursive neural model, namely RST-Recursive, which takes advantage of the text's RST features produced by a state of the art RST parser. We evaluate our approach on the Grammarly Corpus for Discourse Coherence (GCDC) and show that when ensembled with the current state of the art, we can achieve the new state of the art accuracy on this benchmark. Furthermore, when deployed alone, RST-Recursive achieves competitive accuracy while having 62% fewer parameters.

Keywords

Cite

@article{arxiv.2009.14463,
  title  = {Neural RST-based Evaluation of Discourse Coherence},
  author = {Grigorii Guz and Peyman Bateni and Darius Muglich and Giuseppe Carenini},
  journal= {arXiv preprint arXiv:2009.14463},
  year   = {2020}
}

Comments

8 pages, 5 figures, to be published in AACL 2020

R2 v1 2026-06-23T18:54:03.654Z