Top-down Discourse Parsing via Sequence Labelling
2021-02-03EACL 2021Code Available1· sign in to hype
Fajri Koto, Jey Han Lau, Timothy Baldwin
Code Available — Be the first to reproduce this paper.
ReproduceCode
- github.com/fajri91/NeuralRST-TopDownOfficialIn paperpytorch★ 11
Abstract
We introduce a top-down approach to discourse parsing that is conceptually simpler than its predecessors (Kobayashi et al., 2020; Zhang et al., 2020). By framing the task as a sequence labelling problem where the goal is to iteratively segment a document into individual discourse units, we are able to eliminate the decoder and reduce the search space for splitting points. We explore both traditional recurrent models and modern pre-trained transformer models for the task, and additionally introduce a novel dynamic oracle for top-down parsing. Based on the Full metric, our proposed LSTM model sets a new state-of-the-art for RST parsing.
Tasks
Benchmark Results
| Dataset | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| RST-DT | LSTM Dynamic | Standard Parseval (Full) | 50.3 | — | Unverified |
| RST-DT | LSTM Static | Standard Parseval (Full) | 49.4 | — | Unverified |
| RST-DT | Transformer (dynamic) | Standard Parseval (Full) | 49.2 | — | Unverified |
| RST-DT | Transformer (static) | Standard Parseval (Full) | 49 | — | Unverified |