Forward Prediction for Physical Reasoning
Open paper
Benchmark
Model
Rank
Results
visual-reasoning-on-phyre-1b-cross
Dec[Joint]1f
#2
AUCCESS: 40.3
visual-reasoning-on-phyre-1b-within
Dec[Joint]1f
#3
AUCCESS: 80.0
Rank counts only results with a code link.