Wals Roberta Sets 37-70.zip -

: Testing if models like RoBERTa or XLM-RoBERTa have "learned" the typological rules of specific languages during pre-training.

: Definite (37A) and Indefinite (38A) article systems. WALS roberta sets 37-70.zip

: Position of tense-aspect affixes (69A) and the morphological imperative (70A). Use Cases for the Dataset : Testing if models like RoBERTa or XLM-RoBERTa

The features in this range are essential for understanding how different languages handle noun and verb structures. : distance contrasts in demonstratives (41A)

: Inclusive/exclusive distinctions (39A–40A), distance contrasts in demonstratives (41A), and third-person pronouns (43A).

World languages with features and coordinates - Dataset Search