Stronger zero-shot ≠ faster adaptation: GR00T N1.5 beats N1.7 in fine-tuning
rahulshivchand · x · 2026-11-11
Stronger zero-shot models are not always more sample-efficient: GR00T N1.7 beats N1.5/N1.6 zero-shot, yet N1.5 consistently adapts faster and achieves higher success at low demo budgets—suggesting leaderboard zero-shot scores don't predict fine-tuning performance.
More from Embodied
- Maestro paper: cheap Gemini 2.5 beats π₀.5 baselines at zero-shot robot control — dineshjayaraman · 2026-11-11
- OpenDDE: open-source all-atom drug discovery engine hits 51% on hard antibody-antigen benchmark — teortaxesTex · 2026-11-11
- Tesla demos Optimus walking inside a world model at CoRL, real policy in control — Scobleizer · 2026-11-11
- NUS Proposes WM-PPO: World-Model-Predicted Successor States Improve PPO Policy Learning — NationalUniversityofSingapore · 2026-11-11
- DexForesight Boosts Dexterous Manipulation Success From 49.4% to 60.5% With Implicit Spatial Foresight — Sharpa-Robotics · 2026-11-11
- NTU Study Finds Time-Indexed Memory Makes VLA Robots Mistake Pauses for Progress — NanyangTechnologicalUniversity · 2026-11-11