Audio codec models make surprisingly good action tokenizers for dexterous robot arms
chris_j_paxton · x · 2026-09-27
- Trossen Robotics showed two of its arms striking a match fully autonomously, achieved by researcher Ahad Jawaid for a CoRL 2026 paper using the previous-generation arms.
- Key insight: audio codec models turn out to be a surprisingly good recipe for action tokenizers. The work will be presented at CoRL in November.
More from Embodied
- Dev: no point getting LASIK in 2026 — AI glasses will bring face-worn superpowers — willcb · 2026-09-27
- Flying cars finally arrive: rider shares first-hand eVTOL takeoff video — chris_j_paxton · 2026-09-27
- NUS Show-Harness: A Semantic Interface Lets VLMs Control Real Robots Zero-Shot — 机器之心 · 2026-09-27
- Opus 5.5 demos nail spatial perception, but tactile sensing remains VLA robotics' frontier — bookwormengr · 2026-09-27
- Tesla FSD miles about to hit 15 billion, a scale no rival matches — Vjeux · 2026-09-27
- Amazon delivery drones overwhelm a Texas suburb, sparking NIMBY backlash — Polymarket · 2026-09-27