schneiderkamplab/dala-ca-audited

DaLA Catalan — audited grammar and spelling 364,492 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 291,792 583,584 validation 36,339 72,678 test 36,361 72,722 Two… See the full description on the dataset page: https://huggingface.co/datasets/schneiderkamplab/dala-ca-audited.

<iframe src="https://silytics.si/embed/item/dec5416c-34f1-47d2-aca8-25d8f0f581d2" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.