Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Real-world cross-view spatial reasoning on STARE
Loading...
35.6
Accuracy
Panoramic + VDrop
23.536
26.668
29.8
32.932
May 26, 2026
Accuracy
Updated 7d ago
Evaluation Results
Method
Method
Links
Accuracy
Panoramic + VDrop
VDrop=✓
2026.05
35.6
Point Matching + VDrop
VDrop=✓
2026.05
34.4
Panoramic
VDrop=✗
2026.05
32.4
Qwen3-VL-8B
VDrop=✗
2026.05
31.6
Qwen3-VL-4B
VDrop=✗
2026.05
29.2
ThinkMorph (24K)
VDrop=✗
2026.05
28.8
BAGEL-Zebra-CoT (182K)
VDrop=✗
2026.05
28.4
BAGEL
VDrop=✗
2026.05
28
Top-down
VDrop=✗
2026.05
28
Text CoT
VDrop=✗
2026.05
26
Top-down + VDrop
VDrop=✓
2026.05
26
No-Think
VDrop=✗
2026.05
24.8
Point Matching
VDrop=✗
2026.05
24
Feedback
Search any
task
Search any
task