Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Human Preference Evaluation on Long Videos 50 60s
Loading...
30
Win Count
Stream-R1
21.68
23.84
26
28.16
May 5, 2026
Win Count
Tie Count
Lose Count
Win Rate
Updated 28d ago
Evaluation Results
Method
Method
Links
Win Count
Tie Count
Lose Count
Win Rate
Stream-R1
Dimension=Dynamic Reas...
2026.05
30
3
17
63
Stream-R1
Dimension=Visual Quali...
2026.05
29
2
19
60
Stream-R1
Dimension=Overall Pref...
2026.05
28
1
21
57
Stream-R1
Dimension=Temporal Con...
2026.05
25
1
24
51
Stream-R1
Dimension=Text-Video A...
2026.05
22
9
18
54.1
Feedback
Search any
task
Search any
task