CEED-VLA: Consistency Vision-Language-Action Model with Early-Exit Decoding
Open paper
Benchmark
Model
Rank
Results
robot-manipulation-on-calvin
CEED-VLA (LLaVA-VLA, action chunk 5, CALVIN ABC→D)
#14
avg. sequence length (D to D): 3.67
Rank counts only results with a code link.