[HUGGINGFACE]score: 0.76

N_0-VTLA Scales Foundation Models with Tactile Tokens

July 25, 2026

N_0-VTLA is a vision-tactile-language-action foundation model trained on the large-scale NeoData dataset to enable fine-grained, contact-rich manipulation. It utilizes a predictive tactile pathway to distill contact patterns for offline policy improvement via tactile-feedback control.

HOW THIS AFFECTS YOU

●

builderThis advances the state of tactile-feedback control for robotics applications requiring fine-grained manipulation.

●

researcherThis establishes a new training recipe for integrating high-resolution tactile data into multimodal foundation models.

read original ↗huggingface.co

DAILY DIGEST

you don't check 9 sources — we do. one email every morning, read in 2 min. free. unsubscribe anytime. privacy

← back to feed