Can We Trust Embodied Agents? Exploring Backdoor Attacks against Embodied LLM-based Decision-Making Systems

• URL: http://arxiv.org/abs/2405.20774v2
• Date: Sat, 05 Oct 2024 05:17:31 GMT
• Title: Can We Trust Embodied Agents? Exploring Backdoor Attacks against Embodied LLM-based Decision-Making Systems
• Authors: Ruochen Jiao, Shaoyuan Xie, Justin Yue, Takami Sato, Lixu Wang, Yixuan Wang, Qi Alfred Chen, Qi Zhu,
• Abstract summary: Large Language Models (LLMs) have shown significant promise in real-world decision-making tasks for embodied AI. LLMs are fine-tuned to leverage their inherent common sense and reasoning abilities while being tailored to specific applications. This fine-tuning process introduces considerable safety and security vulnerabilities, especially in safety-critical cyber-physical systems.
• Score: 27.316115171846953
• License:
• Abstract: Large Language Models (LLMs) have shown significant promise in real-world decision-making tasks for embodied artificial intelligence, especially when fine-tuned to leverage their inherent common sense and reasoning abilities while being tailored to specific applications. However, this fine-tuning process introduces considerable safety and security vulnerabilities, especially in safety-critical cyber-physical systems. In this work, we propose the first comprehensive framework for Backdoor Attacks against LLM-based Decision-making systems (BALD) in embodied AI, systematically exploring the attack surfaces and trigger mechanisms. Specifically, we propose three distinct attack mechanisms: word injection, scenario manipulation, and knowledge injection, targeting various components in the LLM-based decision-making pipeline. We perform extensive experiments on representative LLMs (GPT-3.5, LLaMA2, PaLM2) in autonomous driving and home robot tasks, demonstrating the effectiveness and stealthiness of our backdoor triggers across various attack channels, with cases like vehicles accelerating toward obstacles and robots placing knives on beds. Our word and knowledge injection attacks achieve nearly 100% success rate across multiple models and datasets while requiring only limited access to the system. Our scenario manipulation attack yields success rates exceeding 65%, reaching up to 90%, and does not require any runtime system intrusion. We also assess the robustness of these attacks against defenses, revealing their resilience. Our findings highlight critical security vulnerabilities in embodied LLM systems and emphasize the urgent need for safeguarding these systems to mitigate potential risks.

Related papers

• Defining and Evaluating Physical Safety for Large Language Models [62.4971588282174]
  Large Language Models (LLMs) are increasingly used to control robotic systems such as drones. Their risks of causing physical threats and harm in real-world applications remain unexplored. We classify the physical safety risks of drones into four categories: (1) human-targeted threats, (2) object-targeted threats, (3) infrastructure attacks, and (4) regulatory violations.
  arXiv  Detail & Related papers  (2024-11-04T17:41:25Z)
• The Best Defense is a Good Offense: Countering LLM-Powered Cyberattacks [2.6528263069045126]
  Large language models (LLMs) could soon become integral to autonomous cyber agents. We introduce novel defense strategies that exploit the inherent vulnerabilities of attacking LLMs. Our results show defense success rates of up to 90%, demonstrating the effectiveness of turning LLM vulnerabilities into defensive strategies.
  arXiv  Detail & Related papers  (2024-10-20T14:07:24Z)
• ASPIRER: Bypassing System Prompts With Permutation-based Backdoors in LLMs [17.853862145962292]
  We introduce a novel backdoor attack that systematically bypasses system prompts. Our method achieves an attack success rate (ASR) of up to 99.50% while maintaining a clean accuracy (CACC) of 98.58%.
  arXiv  Detail & Related papers  (2024-10-05T02:58:20Z)
• A Study on Prompt Injection Attack Against LLM-Integrated Mobile Robotic Systems [4.71242457111104]
  Large Language Models (LLMs) can process multi-modal prompts, enabling them to generate more context-aware responses. One of the primary concerns is the potential security risks associated with using LLMs in robotic navigation tasks. This study investigates the impact of prompt injections on mobile robot performance in LLM-integrated systems.
  arXiv  Detail & Related papers  (2024-08-07T02:48:22Z)
• Security Matrix for Multimodal Agents on Mobile Devices: A Systematic and Proof of Concept Study [16.559272781032632]
  The rapid progress in the reasoning capability of the Multi-modal Large Language Models has triggered the development of autonomous agent systems on mobile devices. Despite the increased human-machine interaction efficiency, the security risks of MLLM-based mobile agent systems have not been systematically studied. This paper highlights the need for security awareness in the design of MLLM-based systems and paves the way for future research on attacks and defense methods.
  arXiv  Detail & Related papers  (2024-07-12T14:30:05Z)
• Prompt Leakage effect and defense strategies for multi-turn LLM interactions [95.33778028192593]
  Leakage of system prompts may compromise intellectual property and act as adversarial reconnaissance for an attacker. We design a unique threat model which leverages the LLM sycophancy effect and elevates the average attack success rate (ASR) from 17.7% to 86.2% in a multi-turn setting. We measure the mitigation effect of 7 black-box defense strategies, along with finetuning an open-source model to defend against leakage attempts.
  arXiv  Detail & Related papers  (2024-04-24T23:39:58Z)
• Physical Backdoor Attack can Jeopardize Driving with Vision-Large-Language Models [53.701148276912406]
  Vision-Large-Language-models (VLMs) have great application prospects in autonomous driving. BadVLMDriver is the first backdoor attack against VLMs for autonomous driving that can be launched in practice using physical objects. BadVLMDriver achieves a 92% attack success rate in inducing a sudden acceleration when coming across a pedestrian holding a red balloon.
  arXiv  Detail & Related papers  (2024-04-19T14:40:38Z)
• Highlighting the Safety Concerns of Deploying LLMs/VLMs in Robotics [54.57914943017522]
  We highlight the critical issues of robustness and safety associated with integrating large language models (LLMs) and vision-language models (VLMs) into robotics applications.
  arXiv  Detail & Related papers  (2024-02-15T22:01:45Z)
• Benchmarking and Defending Against Indirect Prompt Injection Attacks on Large Language Models [82.98081731588717]
  Integration of large language models with external content exposes applications to indirect prompt injection attacks. We introduce the first benchmark for indirect prompt injection attacks, named BIPIA, to evaluate the risk of such attacks. We develop two black-box methods based on prompt learning and a white-box defense method based on fine-tuning with adversarial training.
  arXiv  Detail & Related papers  (2023-12-21T01:08:39Z)
• Unveiling Vulnerabilities in Interpretable Deep Learning Systems with Query-Efficient Black-box Attacks [16.13790238416691]
  Interpretable Deep Learning Systems (IDLSes) are designed to make the system more transparent and explainable. We propose a novel microbial genetic algorithm-based black-box attack against IDLSes that requires no prior knowledge of the target model and its interpretation model.
  arXiv  Detail & Related papers  (2023-07-21T21:09:54Z)
• Invisible for both Camera and LiDAR: Security of Multi-Sensor Fusion based Perception in Autonomous Driving Under Physical-World Attacks [62.923992740383966]
  We present the first study of security issues of MSF-based perception in AD systems. We generate a physically-realizable, adversarial 3D-printed object that misleads an AD system to fail in detecting it and thus crash into it. Our results show that the attack achieves over 90% success rate across different object types and MSF.
  arXiv  Detail & Related papers  (2021-06-17T05:11:07Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.