RoboAlign: Learning Test-Time Reasoning for Language-Action Alignment in Vision-Language-Action Models
Open paper
Benchmark
Model
Rank
Results
robot-manipulation-on-libero
RoboAlign
–
Success Rate: 86.8
zero-shot-generalization-on-calvin
RoboAlign
–
Avg. sequence length: 2.57
Rank counts only results with a code link.