| language: | |
| - en | |
| license: mit | |
| pretty_name: ling_conversion | |
| # ling_conversion | |
| Processed dataset used in the LingConv paper: [Linguistically-Controlled Paraphrase Generation](https://aclanthology.org/2025.findings-emnlp.1137/). | |
| ## Splits | |
| - train | |
| - dev | |
| - test | |
| ## Columns | |
| - sentence1 | |
| - sentence2 | |
| - source | |
| - sentence1_lftk+ | |
| - sentence2_lftk+ | |
| - sentence1_discr | |
| - sentence2_discr | |
| - sentence1_ling | |
| - sentence2_ling | |
| ## Preprocessing | |
| The preprocessing scripts used to build this release are included in `preprocessing/`. | |