BIG-bench Medical Questions Russian tests yes/no comprehension of Russian medical text.
unassessed
| Category | domain |
|---|---|
| Subcategory | Russian medical question answering |
| Page status | active |
| Metric | multiple_choice_grade |
| Direction | higher_is_better |
| Unit | percent |
| Dataset size | 256 |
| Publisher | BIG-bench / Google Research |
The task gives a Russian medical context and a clarifying question. The model must decide whether the question is supported by the context and answer да or нет.
Russian passage and question; binary multiple-choice answer using да or нет.
No model card in ModelSpec reports this benchmark yet.