LaST-R1: Reinforcing Action via Adaptive Physical Latent Reasoning for VLA Models
Project release for LaST-R1, a VLA framework using latent physical reasoning before action generation.
Evidence notes
- Combines adaptive reasoning steps with reinforcement-style post-training for manipulation policies.
- Targets improved generalization and policy refinement across single-arm and dual-arm robot tasks.
Company context
Chinese embodied-robotics company developing general-purpose robot hardware, models and deployment systems.