Search papers, labs, and topics across Lattice.
The Ohio State University
2
0
5
Text-based speculative decoding falls flat for vision-language models, but ViSkip dynamically adapts to vision tokens for state-of-the-art acceleration.
Forget fine-tuning: DM0 shows that pretraining a VLA model from scratch on diverse embodied and non-embodied data leads to SOTA performance in physical AI tasks.
