Figure ablation: Index pretraining lifts robot zero-shot success from 8% to 56%
coreylynch · x · 2026-09-18
Figure's Corey Lynch shows predictable scaling for Helix: doubling Index data improved next-action prediction enough to forecast the largest run's loss to four decimal places before training. A key ablation — same model, data, and eval — shows Index pretraining boosts zero-shot success from 8% to 56%, directly transferring whole-body behaviors.
Related event: Figure Unveils Helix 2.5: Zero-Shot Household Work Across 30 Unseen Homes(27 posts)→
More from Embodied
- Autonomous's Harness Pro uses an ESP32-P4, not a mini PC, to monitor AI coding agents — dee_hw · 2026-11-02
- KAIST's QCP lets robot policies control execution quality in real time — kaist-ai · 2026-11-02
- ANYbotics begins customer deliveries of ANYmal X, first legged robot certified for explosive atmospheres — lukas_m_ziegler · 2026-11-02
- InSpatio opens China's first embodied simulation data center targeting 1M hours/month — 智东西 · 2026-11-02
- GR-BC mines better trajectories from imperfect demos, hitting ~91% success on precise manipulation — 机器之心 · 2026-11-02
- RAM prices tripled, so Tesla cut a third of memory from Optimus's brain — aakashgupta · 2026-11-02