FlexLLM: Exploring LLM Customization for Moving Target Defense on Black-Box LLMs Against Jailbreak Attacks

Defense in large language models (LLMs) is crucial to counter the numerous attackers exploiting these systems to generate harmful content through manipulated prompts, known as jailbreak attacks. Although many defense strategies have been proposed, they often require access to the model's internal structure or need additional training, which is impractical for service providers using LLM APIs, such as OpenAI APIs or Claude APIs. In this paper, we propose a moving target defense approach that alters decoding hyperparameters to enhance model robustness against various jailbreak attacks. Our approach does not require access to the model's internal structure and incurs no additional training costs. The proposed defense includes two key components: (1) optimizing the decoding strategy by identifying and adjusting decoding hyperparameters that influence token generation probabilities, and (2) transforming the decoding hyperparameters and model system prompts into dynamic targets, which are continuously altered during each runtime. By continuously modifying decoding strategies and prompts, the defense effectively mitigates the existing attacks. Our results demonstrate that our defense is the most effective against jailbreak attacks in three of the models tested when using LLMs as black-box APIs. Moreover, our defense offers lower inference costs and maintains comparable response quality, making it a potential layer of protection when used alongside other defense methods.

Motion-guided small MAV detection in complex and non-planar scenes

In recent years, there has been a growing interest in the visual detection of micro aerial vehicles (MAVs) due to its importance in numerous applications. However, the existing methods based on either appearance or motion features encounter difficulties when the background is complex or the MAV is too small. In this paper, we propose a novel motion-guided MAV detector that can accurately identify small MAVs in complex and non-planar scenes. This detector first exploits a motion feature enhancement module to capture the motion features of small MAVs. Then it uses multi-object tracking and trajectory filtering to eliminate false positives caused by motion parallax. Finally, an appearance-based classifier and an appearance-based detector that operates on the cropped regions are used to achieve precise detection results. Our proposed method can effectively and efficiently detect extremely small MAVs from dynamic and complex backgrounds because it aggregates pixel-level motion features and eliminates false positives based on the motion and appearance features of MAVs. Experiments on the ARD-MAV dataset demonstrate that the proposed method could achieve high performance in small MAV detection under challenging conditions and outperform other state-of-the-art methods across various metrics

Protecting Activity Sensing Data Privacy Using Hierarchical Information Dissociation

Smartphones and wearable devices have been integrated into our daily lives, offering personalized services. However, many apps become overprivileged as their collected sensing data contains unnecessary sensitive information. For example, mobile sensing data could reveal private attributes (e.g., gender and age) and unintended sensitive features (e.g., hand gestures when entering passwords). To prevent sensitive information leakage, existing methods must obtain private labels and users need to specify privacy policies. However, they only achieve limited control over information disclosure. In this work, we present Hippo to dissociate hierarchical information including private metadata and multi-grained activity information from the sensing data. Hippo achieves fine-grained control over the disclosure of sensitive information without requiring private labels. Specifically, we design a latent guidance-based diffusion model, which generates multi-grained versions of raw sensor data conditioned on hierarchical latent activity features. Hippo enables users to control the disclosure of sensitive information in sensing data, ensuring their privacy while preserving the necessary features to meet the utility requirements of applications. Hippo is the first unified model that achieves two goals: perturbing the sensitive attributes and controlling the disclosure of sensitive information in mobile sensing data. Extensive experiments show that Hippo can anonymize personal attributes and transform activity information at various resolutions across different types of sensing data.

The Dark Side of Human Feedback: Poisoning Large Language Models via User Inputs

Large Language Models (LLMs) have demonstrated great capabilities in natural language understanding and generation, largely attributed to the intricate alignment process using human feedback. While alignment has become an essential training component that leverages data collected from user queries, it inadvertently opens up an avenue for a new type of user-guided poisoning attacks. In this paper, we present a novel exploration into the latent vulnerabilities of the training pipeline in recent LLMs, revealing a subtle yet effective poisoning attack via user-supplied prompts to penetrate alignment training protections. Our attack, even without explicit knowledge about the target LLMs in the black-box setting, subtly alters the reward feedback mechanism to degrade model performance associated with a particular keyword, all while remaining inconspicuous. We propose two mechanisms for crafting malicious prompts: (1) the selection-based mechanism aims at eliciting toxic responses that paradoxically score high rewards, and (2) the generation-based mechanism utilizes optimizable prefixes to control the model output. By injecting 1\% of these specially crafted prompts into the data, through malicious users, we demonstrate a toxicity score up to two times higher when a specific trigger word is used. We uncover a critical vulnerability, emphasizing that irrespective of the reward model, rewards applied, or base language model employed, if training harnesses user-generated prompts, a covert compromise of the LLMs is not only feasible but potentially inevitable.

SwitchTab: Switched Autoencoders Are Effective Tabular Learners

Self-supervised representation learning methods have achieved significant success in computer vision and natural language processing, where data samples exhibit explicit spatial or semantic dependencies. However, applying these methods to tabular data is challenging due to the less pronounced dependencies among data samples. In this paper, we address this limitation by introducing SwitchTab, a novel self-supervised method specifically designed to capture latent dependencies in tabular data. SwitchTab leverages an asymmetric encoder-decoder framework to decouple mutual and salient features among data pairs, resulting in more representative embeddings. These embeddings, in turn, contribute to better decision boundaries and lead to improved results in downstream tasks. To validate the effectiveness of SwitchTab, we conduct extensive experiments across various domains involving tabular data. The results showcase superior performance in end-to-end prediction tasks with fine-tuning. Moreover, we demonstrate that pre-trained salient embeddings can be utilized as plug-and-play features to enhance the performance of various traditional classification methods (e.g., Logistic Regression, XGBoost, etc.). Lastly, we highlight the capability of SwitchTab to create explainable representations through visualization of decoupled mutual and salient features in the latent space.

Global-Local MAV Detection under Challenging Conditions based on Appearance and Motion

Visual detection of micro aerial vehicles (MAVs) has received increasing research attention in recent years due to its importance in many applications. However, the existing approaches based on either appearance or motion features of MAVs still face challenges when the background is complex, the MAV target is small, or the computation resource is limited. In this paper, we propose a global-local MAV detector that can fuse both motion and appearance features for MAV detection under challenging conditions. This detector first searches MAV target using a global detector and then switches to a local detector which works in an adaptive search region to enhance accuracy and efficiency. Additionally, a detector switcher is applied to coordinate the global and local detectors. A new dataset is created to train and verify the effectiveness of the proposed detector. This dataset contains more challenging scenarios that can occur in practice. Extensive experiments on three challenging datasets show that the proposed detector outperforms the state-of-the-art ones in terms of detection accuracy and computational efficiency. In particular, this detector can run with near real-time frame rate on NVIDIA Jetson NX Xavier, which demonstrates the usefulness of our approach for real-world applications. The dataset is available at https://github.com/WestlakeIntelligentRobotics/GLAD. In addition, A video summarizing this work is available at https://youtu.be/Tv473mAzHbU.

Beyond Boundaries: A Comprehensive Survey of Transferable Attacks on AI Systems

Artificial Intelligence (AI) systems such as autonomous vehicles, facial recognition, and speech recognition systems are increasingly integrated into our daily lives. However, despite their utility, these AI systems are vulnerable to a wide range of attacks such as adversarial, backdoor, data poisoning, membership inference, model inversion, and model stealing attacks. In particular, numerous attacks are designed to target a particular model or system, yet their effects can spread to additional targets, referred to as transferable attacks. Although considerable efforts have been directed toward developing transferable attacks, a holistic understanding of the advancements in transferable attacks remains elusive. In this paper, we comprehensively explore learning-based attacks from the perspective of transferability, particularly within the context of cyber-physical security. We delve into different domains -- the image, text, graph, audio, and video domains -- to highlight the ubiquitous and pervasive nature of transferable attacks. This paper categorizes and reviews the architecture of existing attacks from various viewpoints: data, process, model, and system. We further examine the implications of transferable attacks in practical scenarios such as autonomous driving, speech recognition, and large language models (LLMs). Additionally, we outline the potential research directions to encourage efforts in exploring the landscape of transferable attacks. This survey offers a holistic understanding of the prevailing transferable attacks and their impacts across different domains.

Optimal Spatial-Temporal Triangulation for Bearing-Only Cooperative Motion Estimation

Vision-based cooperative motion estimation is an important problem for many multi-robot systems such as cooperative aerial target pursuit. This problem can be formulated as bearing-only cooperative motion estimation, where the visual measurement is modeled as a bearing vector pointing from the camera to the target. The conventional approaches for bearing-only cooperative estimation are mainly based on the framework distributed Kalman filtering (DKF). In this paper, we propose a new optimal bearing-only cooperative estimation algorithm, named spatial-temporal triangulation, based on the method of distributed recursive least squares, which provides a more flexible framework for designing distributed estimators than DKF. The design of the algorithm fully incorporates all the available information and the specific triangulation geometric constraint. As a result, the algorithm has superior estimation performance than the state-of-the-art DKF algorithms in terms of both accuracy and convergence speed as verified by numerical simulation. We rigorously prove the exponential convergence of the proposed algorithm. Moreover, to verify the effectiveness of the proposed algorithm under practical challenging conditions, we develop a vision-based cooperative aerial target pursuit system, which is the first of such fully autonomous systems so far to the best of our knowledge.

MASTERKEY: Practical Backdoor Attack Against Speaker Verification Systems

Speaker Verification (SV) is widely deployed in mobile systems to authenticate legitimate users by using their voice traits. In this work, we propose a backdoor attack MASTERKEY, to compromise the SV models. Different from previous attacks, we focus on a real-world practical setting where the attacker possesses no knowledge of the intended victim. To design MASTERKEY, we investigate the limitation of existing poisoning attacks against unseen targets. Then, we optimize a universal backdoor that is capable of attacking arbitrary targets. Next, we embed the speaker's characteristics and semantics information into the backdoor, making it imperceptible. Finally, we estimate the channel distortion and integrate it into the backdoor. We validate our attack on 6 popular SV models. Specifically, we poison a total of 53 models and use our trigger to attack 16,430 enrolled speakers, composed of 310 target speakers enrolled in 53 poisoned models. Our attack achieves 100% attack success rate with a 15% poison rate. By decreasing the poison rate to 3%, the attack success rate remains around 50%. We validate our attack in 3 real-world scenarios and successfully demonstrate the attack through both over-the-air and over-the-telephony-line scenarios.

PhantomSound: Black-Box, Query-Efficient Audio Adversarial Attack via Split-Second Phoneme Injection

In this paper, we propose PhantomSound, a query-efficient black-box attack toward voice assistants. Existing black-box adversarial attacks on voice assistants either apply substitution models or leverage the intermediate model output to estimate the gradients for crafting adversarial audio samples. However, these attack approaches require a significant amount of queries with a lengthy training stage. PhantomSound leverages the decision-based attack to produce effective adversarial audios, and reduces the number of queries by optimizing the gradient estimation. In the experiments, we perform our attack against 4 different speech-to-text APIs under 3 real-world scenarios to demonstrate the real-time attack impact. The results show that PhantomSound is practical and robust in attacking 5 popular commercial voice controllable devices over the air, and is able to bypass 3 liveness detection mechanisms with >95% success rate. The benchmark result shows that PhantomSound can generate adversarial examples and launch the attack in a few minutes. We significantly enhance the query efficiency and reduce the cost of a successful untargeted and targeted adversarial attack by 93.1% and 65.5% compared with the state-of-the-art black-box attacks, using merely ~300 queries (~5 minutes) and ~1,500 queries (~25 minutes), respectively.