Sign inGet started
Dataset · Hugging Face

schneiderkamplab/dala-ro-audited

DaLA Romanian — audited grammar and spelling 299,904 synthetic original/corrupted pairs, selected by an automated Gemma audit. Not native-speaker validated or a gold benchmark. No paraphrasing, simplification or style-transfer task. Valid variants should not be labelled errors; model judgments remain fallible and systematic errors can survive filtering. Split Pairs Rows per task train 241,194 482,388 validation 29,354 58,708 test 29,356 58,712 Two… See the full description on the dataset page: https://huggingface.co/datasets/schneiderkamplab/dala-ro-audited.

Author / organizationschneiderkamplab
License (as declared)other
Downloads505 (as reported by Hugging Face)
Likes0
Created28 September 2026
Last updated at source28 September 2026
Embed this entry
<iframe src="https://silytics.si/embed/item/0f45e7b4-a710-4da8-b77f-926296ffd1cb" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Free, no tracking. More about feeds and embeds.

Popularity, last 4 days

505 now. Counts as reported by the source, snapshotted daily by Silytics.

More from schneiderkamplab

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 19 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.