A Unified Linear-Time Framework for Sentence-Level Discourse Parsing
This work addresses efficient discourse analysis for natural language processing, offering incremental improvements in parsing accuracy.
The paper tackles sentence-level discourse parsing by proposing a neural framework with a segmenter and parser based on Pointer Networks, achieving F1 scores of 95.4 for segmentation and 81.7 for parsing, surpassing previous methods and approaching human agreement.
We propose an efficient neural framework for sentence-level discourse analysis in accordance with Rhetorical Structure Theory (RST). Our framework comprises a discourse segmenter to identify the elementary discourse units (EDU) in a text, and a discourse parser that constructs a discourse tree in a top-down fashion. Both the segmenter and the parser are based on Pointer Networks and operate in linear time. Our segmenter yields an $F_1$ score of 95.4, and our parser achieves an $F_1$ score of 81.7 on the aggregated labeled (relation) metric, surpassing previous approaches by a good margin and approaching human agreement on both tasks (98.3 and 83.0 $F_1$).