Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Geometric Reasoning on Geometry3K (avg@8 accuracy)
Loading...
44.11
Avg@8 Accuracy
PAPO_D
28.1044
32.2597
36.415
40.5703
Jul 8, 2025
Avg@8 Accuracy
Updated 3d ago
Evaluation Results
Method
Method
Links
Avg@8 Accuracy
PAPO_D
Backbone=Qwen2.5-VL, M...
2025.07
44.11
PAPO_G
Backbone=Qwen3-VL (thi...
2025.07
41.08
PAPO_G
Backbone=Qwen2.5-VL, M...
2025.07
40.25
GRPO
Backbone=Qwen2.5-VL, M...
2025.07
40.18
GRPO
Backbone=Qwen3-VL (thi...
2025.07
39.29
DAPO
Backbone=Qwen2.5-VL, M...
2025.07
35.92
PAPO_D
Backbone=Qwen2.5-VL, M...
2025.07
35.65
DAPO
Backbone=Qwen2.5-VL, M...
2025.07
31.2
PAPO_G
Backbone=Qwen2.5-VL, M...
2025.07
30.95
GRPO
Backbone=Qwen2.5-VL, M...
2025.07
28.72
Feedback
Search any
task
Search any
task