SafeStage: Evaluating Safety Before, During, and After Vision-Language-Conditioned Robot Manipulation

arXiv:2609.21223 · cs.RO · Submitted 2026-09-18 · Read on arXiv

cs.RO

Submitted: 2026-09-18

Updated: 2026-09-18

Code: https://github.com/JinzhuLuo/SafeStage

Terminology

• Benchmark — 25× in this paper · explained in 2 episodes
• Violation Rate — 7× in this paper
• Execution-Time Safety — 5× in this paper
• Final-State Hazards — 5× in this paper
• Initial-State Hazards — 4× in this paper
• Action Space — 2× in this paper · explained in 1 episode
• Benchmarking — 2× in this paper · explained in 4 episodes
• Fidelity — 2× in this paper · explained in 2 episodes
• Generalization — 2× in this paper · explained in 13 episodes
• Trace — 2× in this paper · explained in 2 episodes
• Transformer — 2× in this paper · explained in 2 episodes
• Alignment — 1× in this paper · explained in 1 episode
• Consistency — 1× in this paper · explained in 3 episodes
• Evaluation Dimensions — 1× in this paper · explained in 3 episodes
• F1 Score — 1× in this paper · explained in 3 episodes
• Failure Modes — 1× in this paper · explained in 3 episodes
• Fine-Tuning — 1× in this paper · explained in 5 episodes
• Foundation Model — 1× in this paper · explained in 3 episodes
• Physical Constraints — 1× in this paper · explained in 7 episodes
• Safe Success Rate — 1× in this paper
• Task Success Rate — 1× in this paper
• Unsafe Success Rate — 1× in this paper
• VLA Models — 1× in this paper · explained in 3 episodes

Sources

• Cosmos 3: Omnimodal World Models for Physical AI
• $\pi_0$: A Vision-Language-Action Flow Model for General Robot Control
• LIBERO-Safety: A Comprehensive Benchmark for Physical and Semantic Safety in Vision-Language-Action Models
• SafeManip: A Property-Driven Benchmark for Temporal Safety Evaluation in Robotic Manipulation
• $\pi_{0.5}$: a Vision-Language-Action Model with Open-World Generalization
• World Action Models are Zero-shot Policies

Related papers

• FMT x: An Efficient and Asymptotically Optimal Extension of the Fast Marching Tree for Dynamic Replanning
• MPCFormer: A physics-informed data-driven approach for explainable socially-aware autonomous driving
• RoboLab: A High-Fidelity Simulation Benchmark for Analysis of Task Generalist Policies
• HRDexDB: A 4D Dexterous Grasping Dataset Across Human and Multiple Robot Embodiments
• APT: Action Expert Pretraining Improves Instruction Generalization of Vision-Language-Action Policies
• Fine-tuning is Not Enough: A Parallel Framework for Collaborative Imitation and Reinforcement Learning in End-to-end Autonomous Driving