MM-Vet
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
MM-Vet
44.7MM-Vet Tool-use Score
13
MM-Vet 58
38.2Score
13
MM-Vet benign queries
54.5Recognition Score
12
MM-VET
39.5REC
12
MM-Vet (5% Forget Set)
43.9Average Score
12
MM-Vet2 (test)
54.4GPT-Eval Score
10
MM-Vet v2
71.8MM-Vet v2 Score
10
MM-Vet
3.82Average Accepted Length (tau)
10
MM-Vet (test)
52.8MM-Vet Score
10
MM-Vet
69.1Accuracy
10
MM-Vet
49.13Utility Score
9
MM-Vet
54.89GPT-4 Score
9
3D MM-Vet (test)
65.1Recognition Accuracy
9
MM-Vet
73.39Pass@5
8
MM-Vet benign
3.21FRR
8
MM-Vet
37.4Score
6
MM-VET v1 (test)
36.4Score
6
MM-Vet (test)
37.7Recognition Accuracy
5
MM-Vet v1 (test)
68.1Utility Score
5
MM-Vet v2 (test)
66.3Utility Score
5
MM-Vet
3.65AIT (s)
5
MM-Vet
3.73Time (s/sample)
5
MM-Vet
57.8Overall Score
4
MM-Vet v2 0613 (test)
71.8Score
4
MM-Vet
44.6MM-Vet Score
4