Silent Sabotage: Internal State Triggered Backdoor Attacks on LLM-Powered Robotic Systems

arXiv:2609.26184 · cs.RO, cs.AI, cs.CR · Submitted 2026-09-02 · Read on arXiv

cs.RO, cs.AI, cs.CR

Submitted: 2026-09-02

Updated: 2026-09-02

DOI: 10.1007/978-3-032-33701-6_7

Code: https://github.com/doniobidov/silent_sabotage

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Fine-Tuning — 6× in this paper · explained in 5 episodes
• Attack Success Rate — 4× in this paper
• Large Language Models (LLMs) — 3× in this paper · explained in 6 episodes
• Attack Success Rate (ASR) — 2× in this paper · explained in 2 episodes
• Core — 2× in this paper · explained in 2 episodes
• In-Context Learning — 2× in this paper · explained in 4 episodes
• In-Context Learning (ICL) — 2× in this paper · explained in 5 episodes
• Instruction Tuning — 2× in this paper · explained in 2 episodes
• Task Completion Rate — 2× in this paper
• Actor — 1× in this paper · explained in 3 episodes
• Affordance — 1× in this paper · explained in 1 episode
• Chain-of-Thought (CoT) — 1× in this paper · explained in 4 episodes
• Chain-of-Thought Prompting — 1× in this paper · explained in 2 episodes
• Embodied AI — 1× in this paper · explained in 2 episodes
• Fidelity — 1× in this paper · explained in 2 episodes
• Generative AI — 1× in this paper · explained in 2 episodes
• Grounding — 1× in this paper · explained in 3 episodes
• Pruning — 1× in this paper · explained in 4 episodes
• Reinforcement Learning — 1× in this paper · explained in 4 episodes
• Vision-Language Model (VLM) — 1× in this paper · explained in 3 episodes

Sources

• BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chain
• Composite Backdoor Attacks Against Large Language Models
• Can We Trust Embodied Agents? Exploring Backdoor Attacks against Embodied LLM-based Decision-Making Systems
• Backdoor Attacks for In-Context Learning with Language Models
• Long-horizon Locomotion and Manipulation on a Quadrupedal Robot with Large Language Models
• Mind the Style of Text! Adversarial and Backdoor Attacks Based on Text Style Transfer
• Hidden Killer: Invisible Textual Backdoor Attacks with Syntactic Trigger
• Turn the Combination Lock: Learnable Textual Backdoor Attacks via Word Substitution
• Universal Jailbreak Backdoors from Poisoned Human Feedback
• BadGPT: Exploring Security Vulnerabilities of ChatGPT via Backdoor Attacks to InstructGPT
• Robot Collapse: Supply Chain Backdoor Attacks Against VLM-based Robotic Manipulation
• BadChain: Backdoor Chain-of-Thought Prompting for Large Language Models
• Instructions as Backdoors: Backdoor Vulnerabilities of Instruction Tuning for Large Language Models
• SafeEmbodAI: a Safety Framework for Mobile Robots in Embodied AI Systems

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