What the Guard Misses, the Robot Executes: Implied Harm in VLA Instructions
cs.RO
Submitted: 2026-10-05
Updated: 2026-10-05
Sources
- Safety Monitors Mostly Catch What the Model Already Refuses
- HazardArena: Evaluating Semantic Safety in Vision-Language-Action Models
- Easier Said Than Done: Unpacking Intent-Behavior Gap in Jailbreaking LLM-Based Robots
- LIBERO-Para: A Diagnostic Benchmark and Metrics for Paraphrase Robustness in VLA Models
- Observing and Controlling Features in Vision-Language-Action Models
- Qwen3Guard Technical Report
- Llama 2: Open Foundation and Fine-Tuned Chat Models
- Gemma: Open Models Based on Gemini Research and Technology
- PaliGemma: A versatile 3B VLM for transfer
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
- ARC: A Reasoning Recipe for Robot Foundation Models
- HRDexDB: A 4D Dexterous Grasping Dataset Across Human and Multiple Robot Embodiments
- APT: Action Expert Pretraining Improves Instruction Generalization of Vision-Language-Action Policies