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

GeoAux-Bench

Benchmarks

Task NameDataset NameSOTA ResultTrend
Geometric ReasoningGeoAux-Bench Core (Total)
Accuracy83.16
21
Geometric ReasoningGeoAux-Bench-Core Olympiad Junior
Accuracy95.17
21
Geometric ReasoningGeoAux-Bench-Core Olympiad Senior
Accuracy89.45
21
Geometric ReasoningGeoAux-Bench-Core Curriculum Junior
Accuracy (%)84.75
21
Geometric ReasoningGeoAux-Bench-Core Curriculum Senior
Accuracy83.91
21
Showing 5 of 5 rows