Recursia hits 28-53% zero-shot on LIBERO-Pro Long, far ahead of ASPIRE and π0.5
jiqizhixin · x · 2026-10-22
- Beijing Jiaotong University, Peking University, NYU, USC, and Beijing Lingxing Intelligence present Recursia, an Embodied Recursive Self-Improvement paradigm for compositional generalization.
- Results: zero-shot success on LIBERO-Pro Long under position/task perturbation is 28% and 53%; NVIDIA's ASPIRE scores 5.8 and 15.1 points lower, OpenVLA and π0 sit near 0%, π0.5 just 8% and 1%.
- Method: a shared executable stack (tools/skills/handoffs/verifications) with failure-driven patching, generality/mechanism/paired-regression verification, and frozen-stack zero-shot target execution.
Related event: Recursia: Embodied Recursive Self-Improvement Shines on LIBERO-Pro(2 posts)→
More from Embodied
- RAX drives robots with a frozen LLM via pure exploration at 1/20th Opus cost — du_yilun · 2026-10-22
- VLA policies bind motor programs to absolute coordinates: injection task succeeds 99.6% on π0.5 — Bryce Grant · 2026-10-22
- Robots will strip human hours out of prices, leaving materials plus energy — Dr_Singularity · 2026-10-22
- NVIDIA open-sources SimReady Central: 2,000+ sim-ready 3D assets on Hugging Face — lhoestq · 2026-10-22
- Dev builds lightweight real-time AI conversational avatars, native in Unreal Engine 5 — Sad_Eagle_937 · 2026-10-22
- Jensen Huang's GTC Berlin finale: one-prompt Hermes Agent builds and trains a robot in simulation — NousResearch · 2026-10-22