Refuses in Chat, Complies with Arms: LLM Refusal Does Not Survive Robot Control
cs.HC, cs.RO
Submitted: 2026-10-03
Updated: 2026-10-03
Code: https://github.com/UKGovernmentBEIS/inspect_ai
Sources
- An Unexpected Robot Policy: Early Evaluations of GPT-6 Astra on RoboDojo and Beyond
- Agent as Policy for Robotic Manipulation
- HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal
- AgentHarm: A Benchmark for Measuring Harmfulness of LLM Agents
- Refusal-Trained LLMs Are Easily Jailbroken As Browser Agents
- MM-SafetyBench: A Benchmark for Safety Evaluation of Multimodal Large Language Models
- LLM-Driven Robots Risk Enacting Discrimination, Violence, and Unlawful Actions
- Jailbreaking LLM-Controlled Robots
- BadRobot: Jailbreaking Embodied LLM Agents in the Physical World
- Code as Policies: Language Model Programs for Embodied Control
- Do As I Can, Not As I Say: Grounding Language in Robotic Affordances
- RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control
- OpenVLA: An Open-Source Vision-Language-Action Model
- $\pi_0$: A Vision-Language-Action Flow Model for General Robot Control
- GR00T N1: An Open Foundation Model for Generalist Humanoid Robots
- MolmoAct: Action Reasoning Models that can Reason in Space
- Gemini Robotics: Bringing AI into the Physical World
- Embodied Red Teaming for Auditing Robotic Foundation Models
- Generating Robot Constitutions & Benchmarks for Semantic Safety
- SafeAgentBench: A Benchmark for Safe Task Planning of Embodied LLM Agents
Related papers
- Calibrating LLM Explanations to Human Decisions with Reinforcement Learning
- EduGage: A Multimodal Dataset and Benchmark for Sensor-Based Momentary Assessment of Engagement in Self-Guided Video Learning
- EvoDesign: Agentic Editable Diagram Creation via Design Expertise Evolution
- HAGI++: Head-Assisted Gaze Imputation and Generation
- Explanation User Interfaces: A Systematic Literature Review
- Hybrid Cinematography: Previsualizing and Managing Hallucination Risk in Generative Video Reshooting