MLVU
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
MLVU
83.8Accuracy
265
MLVU
78.19Score
233
MLVU
76.2Accuracy
213
MLVU
81.5Accuracy
147
MLVU
76.2M-Avg Score
80
MLVU
87.34Accuracy
80
MLVU (dev)
78.1Score
63
MLVU (test)
81Average Score
60
MLVU
68.9Score
55
MLVU 3-120min (test)
47.7Accuracy
49
MLVU 3-120min (dev)
63Accuracy
49
MLVU
79.5M-Avg
46
MLVU 78 (test)
76.66Accuracy
45
MLVU 3-120 min
82.1Accuracy
36
MLVU (dev)
78.1Accuracy
34
MLVU (test)
100.3Average
34
MLVU
102.8Accuracy
33
MLVU v1.0 (test)
67.77MLVU Score
28
MLVU v1 (test)
71.5Accuracy
26
MLVU 1.0 (test)
68.5Accuracy
26
MLVU (dev)
70.2MLVU Dev Score
24
MLVU MCQ (test)
81.5Accuracy
22
MLVU multiple-choice task
73.4Overall Accuracy
21
MLVU MCQ (test)
81.5Accuracy
21
MLVU
73.46Accuracy
21