Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Step recognition on COIN
Loading...
67.3
Top-1 Accuracy
GAD
2.3
19.175
36.05
52.925
Mar 3, 2026
Top-1 Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Top-1 Accuracy
GAD
Backbone=Llama3-8B-Ins...
2026.03
67.3
GAD
Evaluation Protocol=fi...
2026.03
67.3
Disc
Backbone=Llama3-8B-Ins...
2026.03
66.4
GAD
Backbone=Llama3.2-1B-I...
2026.03
65.3
Disc
Backbone=Llama3.2-1B-I...
2026.03
64.1
StreamMind-8B
Backbone=8B
2026.03
63.7
Videollm-MoD-8B
Backbone=8B
2026.03
63.4
Videollm-online-8B
Backbone=8B
2026.03
63.1
VideoTaskGraph
2026.03
57.2
Qwen2.5-VL-7B
Evaluation Protocol=op...
2026.03
16.1
Qwen2.5-VL-7B
Prompting Strategy=cat...
2026.03
11.9
Videollm-online-8B
Evaluation Protocol=op...
2026.03
4.8
Feedback
Search any
task
Search any
task