5,000 hours of paid touch data: robotics' scarcest, unscrapeable training input
kimmonismus · x · 2026-11-11
Part 3 explains the data moat behind LingBot-Touch:
- Raw sensor signals gave at most about half the gain of learned touch tokens.
- Across Diffusion Policy, π0.5, LingBot-VLA and T-Rex, learned touch tokens added 14–25 percentage points; best setup in 89 of 104 task-policy pairs.
- Robbyant recorded 5,000 hours / 1.33B frames of touch data from tactile gloves and teleoperated hands — data nobody uploads for free, making touch one of robotics' scarcest inputs.
Related event: LingBot-Touch: tactile tokens boost robot policy success by 14-25 points(3 posts)→
More from Embodied
- CoRL best-paper nominee Play2Perfect ported to new robots in IsaacLab and MJLab — leto__jean · 2026-11-11
- Walden Robotics lays out its embodied AI playbook: full-stack, sim, teleop, human data — chris_j_paxton · 2026-11-11
- One touch encoder generalizes across sensors, cutting contact error to 0.16mm — kimmonismus · 2026-11-11
- Chris Paxton rides a Zoox robotaxi for the first time since his internship 10 years ago — chris_j_paxton · 2026-11-11
- Robotics VC bull case: Physical AI funding still below pre-ChatGPT levels, winners will win — Rewkang · 2026-11-11
- Binary-reward RL hits 94% success on hard robotics task, recipe looks clear — kastnerkyle · 2026-11-11