A Simple and Strong Baseline for End-to-End Neural RST-style Discourse Parsing
This provides a reference baseline for researchers in discourse parsing, though it is incremental as it combines existing methods.
The paper tackles the need for a strong baseline in RST-style discourse parsing by integrating simple parsing strategies with transformer-based language models, finding that performance depends more on the language model than the parsing strategy, with DeBERTa achieving large gains over the current best parser.
To promote and further develop RST-style discourse parsing models, we need a strong baseline that can be regarded as a reference for reporting reliable experimental results. This paper explores a strong baseline by integrating existing simple parsing strategies, top-down and bottom-up, with various transformer-based pre-trained language models. The experimental results obtained from two benchmark datasets demonstrate that the parsing performance strongly relies on the pretrained language models rather than the parsing strategies. In particular, the bottom-up parser achieves large performance gains compared to the current best parser when employing DeBERTa. We further reveal that language models with a span-masking scheme especially boost the parsing performance through our analysis within intra- and multi-sentential parsing, and nuclearity prediction.