Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Demonstration Selection on Demo-ICL-Bench
Loading...
76
Overall Accuracy
Human
12.04
28.645
45.25
61.855
Feb 9, 2026
Overall Accuracy
Selection Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Overall Accuracy
Selection Accuracy
Human
2026.02
76
88
Gemini-2.5-Pro
2026.02
26
-
GPT-4o
2026.02
24.5
-
Demo-ICL
Size=7B, Frame=32
2026.02
24
58
Demo-ICL (SFT)
Size=7B, Frame=32
2026.02
21.5
54.5
LLaVA-Video
Size=7B, Frame=32
2026.02
20.5
44.5
VideoChat-R1
Size=7B, Frame=32
2026.02
18.5
52
Qwen2.5-VL
Size=7B, Frame=32
2026.02
18
46
Qwen2.5-VL
Size=72B, Frame=32
2026.02
18
54
Ola-Video (Base)
Size=7B, Frame=32
2026.02
18
48
Video-R1
Size=7B, Frame=32
2026.02
17.5
48
InternVL-3
Size=8B, Frame=32
2026.02
16.5
44
Ola
Size=7B, Frame=32
2026.02
16
43
Qwen2-VL
Size=7B, Frame=32
2026.02
14.5
38
Feedback
Search any
task
Search any
task