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.

利用 Rhetorical Structure Theory (RST) 树和关系对话语连贯性进行评估，表明在分类连贯性方面增加银标准 RST 特征可以提高准确性，我们开发了一个 tree-recursive neural model，即 RST-Recursive，利用了由最先进的 RST 解析器产生的文本 RST 特征。在 Grammarly Corpus for Discourse Coherence (GCDC) 上评估我们的方法，并展示了当与现有先进技术结合使用时，我们可以在此基准测试中实现新的技术水平的准确性。此外，单独部署时，RST-Recursive 具有竞争性的准确性，同时参数减少了 62%。