Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multimodal Math Reasoning on MathVista (avg@8 accuracy)
Loading...
69.53
Avg@8 Accuracy
PAPO_G
52.942
57.2485
61.555
65.8615
Jul 8, 2025
Avg@8 Accuracy
Updated 3d ago
Evaluation Results
Method
Method
Links
Avg@8 Accuracy
PAPO_G
Backbone=Qwen2.5-VL, M...
2025.07
69.53
PAPO_D
Backbone=Qwen2.5-VL, M...
2025.07
67.53
GRPO
Backbone=Qwen2.5-VL, M...
2025.07
65.48
PAPO_D
Backbone=Qwen2.5-VL, M...
2025.07
62.53
DAPO
Backbone=Qwen2.5-VL, M...
2025.07
61.91
PAPO_G
Backbone=Qwen2.5-VL, M...
2025.07
61.38
DAPO
Backbone=Qwen2.5-VL, M...
2025.07
60.89
GRPO
Backbone=Qwen2.5-VL, M...
2025.07
59.34
PAPO_G
Backbone=Qwen3-VL (thi...
2025.07
56.08
GRPO
Backbone=Qwen3-VL (thi...
2025.07
53.58
Feedback
Search any
task
Search any
task