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

STI-Bench

Benchmarks

Task NameDataset NameSOTA ResultTrend
Spatial ReasoningSTI-Bench
D-Measure Score37.5
18
Video Spatial ReasoningSTI-Bench
Accuracy41.4
15
3D/4D Video Question AnsweringSTI-Bench
Accuracy41.4
12
3D/4D Visual Question AnsweringSTI Bench 1.0 (test)
Accuracy59.1
8
Showing 4 of 4 rows