|
--- |
|
language: |
|
- en |
|
tags: |
|
- discourse-marker-prediction |
|
- discourse-connective-prediction |
|
- discourse-connective |
|
- discourse-marker |
|
- discourse-relation-prediction |
|
- pragmatics |
|
- discourse |
|
license: apache-2.0 |
|
datasets: |
|
- discovery |
|
metrics: |
|
- accuracy |
|
widget: |
|
- text: "But no, Amazon selling 3D printers is not new.</s></s>The promise of 3D printing is very great." |
|
--- |
|
|
|
# Discourse marker prediction / discourse connective prediction pretrained model |
|
|
|
`roberta-base` pretrained on discourse marker prediction on the Discovery dataset with a validation accuracy of 30.93% (majority class is 0.57%) |
|
|
|
https://github.com/sileod/discovery |
|
|
|
https://huggingface.co/datasets/discovery |
|
|
|
This model can also be used as a pretrained model for NLU, pragmatics and discourse tasks |
|
|
|
## Citing & Authors |
|
|
|
```bibtex |
|
@inproceedings{sileo-etal-2019-mining, |
|
title = "Mining Discourse Markers for Unsupervised Sentence Representation Learning", |
|
author = "Sileo, Damien and |
|
Van De Cruys, Tim and |
|
Pradel, Camille and |
|
Muller, Philippe", |
|
booktitle = "Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)", |
|
month = jun, |
|
year = "2019", |
|
address = "Minneapolis, Minnesota", |
|
publisher = "Association for Computational Linguistics", |
|
url = "https://aclanthology.org/N19-1351", |
|
doi = "10.18653/v1/N19-1351", |
|
pages = "3477--3486", |
|
} |
|
``` |