MV-STRIDE: Enabling MLLMs to Master Multi-View Spatial Reasoning via Hierarchical Capability Modeling
Open paper
Benchmark
Model
Rank
Results
multimodal-reasoning-on-cv-bench
MV-STRIDE-SFT (Full-data SFT) (extra data)
#3
Accuracy: 86.73
spatial-reasoning-on-mmsi-bench
MV-STRIDE-SFT (Full-data SFT) (extra data)
#3
Accuracy: 38.9
Rank counts only results with a code link.