Add link to paper and improve dataset card

Update README.md

Update README.md

Update README.md

Update README.md

Update README.md

Update README.md

fix: per-task subsets + train/test splits for Dataset Viewer

fix: per-task subsets + train/test splits for Dataset Viewer

fix: per-task subsets + train/test splits for Dataset Viewer

GENEB: 100 tasks as subsets with train/test splits

GENEB: 100 tasks as subsets with train/test splits

GENEB benchmark: 100 frozen task CSVs (text,label,split)

initial commit