HPLT/DocHPLT
DocHPLT: A Massively Multilingual Document-Level Translation Dataset Existing document-level machine translation resources are only available for a handful of languages, mostly high-resourced ones. To facilitate the training and evaluation of document-level translation and, more broadly, long-context modeling for global communities, we create DocHPLT, the largest publicly available document-level translation dataset to date. It contains 124 million aligned document pairs across… See the full description on the dataset page: https://huggingface.co/datasets/HPLT/DocHPLT.
| Author / organization | HPLT |
|---|---|
| License (as declared) | cc0-1.0 |
| Downloads | 4.7K (as reported by Hugging Face) |
| Likes | 21 |
| Created | 9 July 2025 |
| Last updated at source | 13 January 2026 |
Popularity, last 2 days
4,688 now (+539 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.
Recent changes
- 16 Oct 2026 Description: DocHPLT: A Massively Multilingual Document-Level Translation Dataset Existing document-level machine translation resources are only available for a handful of languages, mostly high-resourced ones. To facilitate the training and evaluation → DocHPLT: A Massively Multilingual Document-Level Translation Dataset Existing document-level machine translation resources are only available for a handful of languages, mostly high-resourced ones. To facilitate the training and evaluation
Changes detected by Silytics between refreshes of the source.
More from HPLT
- HPLT/HPLT2.0_cleanedDataset
- HPLT/HPLT3.0Dataset
- HPLT/DocHPLT-v3-cleanDataset
Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.