DeeR-VLA: Dynamic Inference of Multimodal Large Language Models for Efficient Robot Execution
Open paper
Benchmark
Model
Rank
Results
robot-manipulation-on-calvin-abcd-to-d
DeeR w. online (CALVIN ABCD→D)
#10
Average Length: 4.13
zero-shot-generalization-on-calvin
DeeR w. online (CALVIN ABC→D)
#14
Avg. sequence length: 2.90
Rank counts only results with a code link.