Discourse Parsing On Rst Dt
Metrics
RST-Parseval (Nuclearity)
RST-Parseval (Relation)
RST-Parseval (Span)
Results
Performance results of various models on this benchmark
Comparison Table
| Model Name | RST-Parseval (Nuclearity) | RST-Parseval (Relation) | RST-Parseval (Span) |
|---|---|---|---|
| a-linear-time-bottom-up-discourse-parser-with | 71.0 | 58.2 | 85.7 |
| a-simple-and-strong-baseline-for-end-to-end | - | - | - |
| can-we-obtain-significant-success-in-rst | - | - | - |
| a-simple-and-strong-baseline-for-end-to-end | - | - | - |
| unleashing-the-power-of-neural-discourse | - | - | - |
| empirical-comparison-of-dependency | 66.6* | 54.6* | 82.6* |
| two-practical-rhetorical-structure-theory | 67.1* | 55.4* | 82.6* |
| a-novel-discourse-parser-based-on-support | 68.4 | 55.3 | 83.0 |
| can-we-obtain-significant-success-in-rst | - | - | - |
| top-down-discourse-parsing-via-sequence | - | - | - |
| a-two-stage-parsing-method-for-text-level | 72.4 | 59.7 | 86.0 |
| a-simple-and-strong-baseline-for-end-to-end | - | - | - |
| bilingual-rhetorical-structure-parsing-with | - | - | - |
| multi-view-and-multi-task-training-of-rst | 63.6* | 47.7* | 79.7* |
| top-down-discourse-parsing-via-sequence | - | - | - |
| can-we-obtain-significant-success-in-rst | - | - | - |
| improving-neural-rst-parsing-model-with | 74.7 | 62.5 | 86.8 |
| a-simple-and-strong-baseline-for-end-to-end | - | - | - |
| rst-parsing-from-scratch | 76.0 | 61.8 | 87.6 |
| a-simple-and-strong-baseline-for-end-to-end | - | - | - |
| top-down-discourse-parsing-via-sequence | - | - | - |
| a-simple-and-strong-baseline-for-end-to-end | - | - | - |
| improving-neural-rst-parsing-model-with | 75.0 | 63.2 | 87.1 |
| can-we-obtain-significant-success-in-rst | - | - | - |
| unleashing-the-power-of-neural-discourse | - | - | - |
| can-we-obtain-significant-success-in-rst | - | - | - |
| unleashing-the-power-of-neural-discourse-1 | - | - | - |
| a-simple-and-strong-baseline-for-end-to-end | - | - | - |
| a-simple-and-strong-baseline-for-end-to-end | - | - | - |
| codra-a-novel-discriminative-framework-for | 68.90 | 55.87 | 83.84 |
| a-simple-and-strong-baseline-for-end-to-end | - | - | - |
| discourse-parsing-with-attention-based | 66.5* | 51.4* | 82.2* |
| top-down-discourse-parsing-via-sequence | - | - | - |
| rst-parsing-from-scratch | - | - | - |
| cross-lingual-rst-discourse-parsing | 68.1* | 56.3* | 81.3* |
| top-down-rst-parsing-utilizing-granularity | 74.6 | 60.0 | 87.0 |
| a-simple-and-strong-baseline-for-end-to-end | - | - | - |
| representation-learning-for-text-level | 68.2* | 57.8* | 82.0* |
| can-we-obtain-significant-success-in-rst | - | - | - |
| transition-based-neural-rst-parsing-with | 73.1 | 60.2 | 85.5 |