VLADrop: Drop-Then-Recovery (DTR) Checkpoints
Collection
Checkpoints for 'Drop-Then-Recovery: How Redundant Are Vision-Language-Action Models?' (arXiv:2606.27755). Code: https://github.com/s1ghhh/VLADrop • 64 items • Updated
VLADrop
Resources for Drop-Then-Recovery (DTR), including VLADrop checkpoints for studying and recovering redundancy in vision-language-action models.