Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Multimodal Reasoning on GeomVerse
Loading...
6.67
Mean@1
Socratic-Solver-Geo (Stage3)
3.1964
4.0982
5
5.9018
Feb 3, 2026
Mean@1
Updated 4d ago
Evaluation Results
Method
Method
Links
Mean@1
Socratic-Solver-Geo (Stage3)
Data Scale=2.5k, Curri...
2026.02
6.67
GeoReasoning
Data Scale=10k
2026.02
5.56
PGPS9K
Data Scale=10k
2026.02
4.44
TrustGeoGen
Data Scale=10k
2026.02
4.44
KD (Our Synthesis)
Data Scale=2.5k
2026.02
3.89
Qwen2.5-VL-7B-Instruct
Mode=Zero-shot
2026.02
3.33
R-CoT
Data Scale=7.2k
2026.02
3.33
Geo170k (G-LLaVA)
Data Scale=10k
2026.02
3.33
KD (Geo3K)
Data Scale=3k
2026.02
3.33
Socratic-Solver-Geo (Stage1)
Data Scale=0.4k, Curri...
2026.02
3.33
Socratic-Solver-Geo (Stage2)
Data Scale=1k, Curricu...
2026.02
3.33
Feedback
Search any
task
Search any
task