pi-StepNFT: Wider Space Needs Finer Steps in Online RL for Flow-based VLAs

arXiv:2603.02083 · cs.RO, cs.CV · Submitted 2026-03-02 · Read on arXiv

cs.RO, cs.CV

Submitted: 2026-03-02

Updated: 2026-10-29

Terminology

• Fine-Tuning — 1× in this paper · explained in 5 episodes
• Generalization — 1× in this paper · explained in 14 episodes
• Online Reinforcement Learning — 1× in this paper · explained in 2 episodes
• Overfitting — 1× in this paper · explained in 2 episodes
• Reinforcement Learning — 1× in this paper · explained in 4 episodes
• Robustness — 1× in this paper · explained in 6 episodes
• Vision-Language-Action (VLA) Models — 1× in this paper · explained in 5 episodes

Related papers

• VLA-Corrector: Lightweight Detect-and-Correct Inference for Adaptive Action Horizon
• FMT x: An Efficient and Asymptotically Optimal Extension of the Fast Marching Tree for Dynamic Replanning
• SafeDojo: Safe Reinforcement Learning for VLA via Interactive World Model
• Feedback World Model Enables Precise Guidance of Diffusion Policy
• Scene Graph-Driven Haptic Feedback for Safety Enhancement in Robotic Ophthalmic Surgery via Physically Simulated iOCT
• ASPIRE: Agentic /Skills Discovery for Robotics