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

SpatialEval

Benchmarks

Task NameDataset NameSOTA ResultTrend
Spatial ReasoningSpatialEval (test)
Maze Navigation Acc35.2
16
Spatial ReasoningSpatialEval
Accuracy70.81
12
Spatial Reasoning (Single-Image)SpatialEval Real
Accuracy68.9
10
Showing 3 of 3 rows