Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Geometry reasoning on Geometry3k (test)
Loading...
48.11
Test Score
DS (Oracle)
40.1332
42.2041
44.275
46.3459
Mar 11, 2026
Test Score
Rollouts
Updated 1mo ago
Evaluation Results
Method
Method
Links
Test Score
Rollouts
DS (Oracle)
Backbone=Qwen2.5-VL-7B...
2026.03
48.11
782
DPS (Ours)
Backbone=Qwen2.5-VL-7B...
2026.03
47.78
328
HR
Backbone=Qwen2.5-VL-7B...
2026.03
46.52
328
US
Backbone=Qwen2.5-VL-7B...
2026.03
46.22
328
DPS (Ours)
Backbone=Qwen2.5-VL-3B...
2026.03
44.47
492
DS (Oracle)
Backbone=Qwen2.5-VL-3B...
2026.03
44.33
1,262
US
Backbone=Qwen2.5-VL-3B...
2026.03
40.69
492
HR
Backbone=Qwen2.5-VL-3B...
2026.03
40.44
492
Feedback
Search any
task
Search any
task