RST Parsing on RST-DT (test)
76.3Span ScoreZhang et al.
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| Zhang et al.Strategy=Top-down, Language Model=XLNet2022.10 | 76.3 | 65.5 | 55.6 | 53.8 | |
| Guz and CareniniStrategy=Bottom-up, Language Model=SpanBERT2022.10 | 75.8 | 64.6 | 53.7 | 51.4 | |
| Nguyen et al.Strategy=Top-down, Language Model=XLNet2022.10 | 74.3 | 64.3 | 51.6 | 50.2 | |
| Kobayashi et al.Strategy=Top-down, Language Model=Glove+ELMo2022.10 | 74.1 | 64.7 | 54.1 | 52.7 | |
| Koto et al.Strategy=Top-down, Language Model=GloVe2022.10 | 73.1 | 62.3 | 51.5 | 50.3 | |
| Guz et al.Strategy=Bottom-up, Language Model=RoBERTa2022.10 | 72.9 | 61.9 | — | — | |
| Wang et al.Strategy=Bottom-up2022.10 | 72 | 60.5 | 50.4 | 48.2 |