Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Blind

Benchmarks

Task NameDataset NameSOTA ResultTrend
Headline GenerationBlind N ≈ 5000 (test)
Mean Rating Improvement (%)2.07
4
Visual ReasoningBlind (test)
Line Intersections Accuracy (Test)90.5
3
Sector classificationBlind
Top-1 Accuracy91.4
2
Showing 3 of 3 rows