The intellectual property of deep generative networks (GNets) can be protected using a cascaded hiding network (HNet) which embeds watermarks (or marks) into GNet outputs, known as box-free watermarking. Although both GNet and HNet are encapsulated in a black box (called operation network, or ONet), with only the generated and marked outputs from HNet being released to end users and deemed secure, in this paper, we reveal an overlooked vulnerability in such systems. Specifically, we show that the hidden GNet outputs can still be reliably estimated via query-based reverse engineering, leaking the generated and unmarked images, despite the attacker's limited knowledge of the system. Our first attempt is to reverse-engineer an inverse model for HNet under the stringent black-box condition, for which we propose to exploit the query process with specially curated input images. While effective, this method yields unsatisfactory image quality. To improve this, we subsequently propose an alternative method leveraging the equivalent additive property of box-free model watermarking and reverse-engineering a forward surrogate model of HNet, with better image quality preservation. Extensive experimental results on image processing and image generation tasks demonstrate that both attacks achieve impressive watermark removal success rates (100%) while also maintaining excellent image quality (reaching the highest PSNR of 34.69 dB), substantially outperforming existing attacks, highlighting the urgent need for robust defensive strategies to mitigate the identified vulnerability in box-free model watermarking.

The development of machine learning models increasingly relies on high-quality data that resides in private domains. To enable secure and value-driven data exchange under strict privacy regulations, federated learning (FL) has emerged as a key primitive by enabling the trading of model utilities instead of raw data. Among existing solutions, martFL (CCS 2023) represents the state-of-the-art FL-based data marketplace architecture, integrating privacy-preserving model evaluation and verifiable trading protocols to enable robust and fair model utility trading without revealing raw data. Despite its strengths, martFL suffers from critical weaknesses at the evaluation layer, including plaintext score exposure and unverifiable and manipulable participant selection. To address these challenges, we propose MartDE, a dedicated evaluation framework that builds model-centric data marketplaces with robust, privacy-preserving, and verifiable mechanisms. MartDE introduces encrypted utility scoring with client-side decryption to preserve score confidentiality, formally bounded anomaly filtering, adaptive participant selection based on global model performance, and commitment-based verification to ensure consistency between declared and evaluated scores and selection verification. We implement MartDE and evaluate it across diverse datasets and adversarial conditions. Results show that MartDE achieves superior accuracy, robustness, and cost-efficiency, providing a strong foundation for secure and trustworthy utility-driven data marketplaces.

Model Context Protocol (MCP) standardizes interface mapping for large language models (LLMs) to access external data and tools, which revolutionizes the paradigm of tool selection and facilitates the rapid expansion of the LLM agent tool ecosystem. However, as the MCP is increasingly adopted, third-party customized versions of the MCP server expose potential security vulnerabilities. In this paper, we first introduce a novel security threat, which we term the MCP Preference Manipulation Attack (MPMA). An attacker deploys a customized MCP server to manipulate LLMs, causing them to prioritize it over other competing MCP servers. This can result in economic benefits for attackers, such as revenue from paid MCP services or advertising income generated from free servers. To achieve MPMA, we first design a Direct Preference Manipulation Attack (DPMA) that achieves significant effectiveness by inserting the manipulative word and phrases into the tool name and description. However, such a direct modification is obvious to users and lacks stealthiness. To address these limitations, we further propose Genetic-based Advertising Preference Manipulation Attack (GAPMA). GAPMA employs four commonly used strategies to initialize descriptions and integrates a Genetic Algorithm (GA) to enhance stealthiness. The experiment results demonstrate that GAPMA balances high effectiveness and stealthiness. Our study reveals a critical vulnerability of the MCP in open ecosystems, highlighting an urgent need for robust defense mechanisms to ensure the fairness of the MCP ecosystem.

Backdoor attacks pose a significant threat to Large Language Models (LLMs), where adversaries can embed hidden triggers to manipulate LLM's outputs. Most existing defense methods, primarily designed for classification tasks, are ineffective against the autoregressive nature and vast output space of LLMs, thereby suffering from poor performance and high latency. To address these limitations, we investigate the behavioral discrepancies between benign and backdoored LLMs in output space. We identify a critical phenomenon which we term sequence lock: a backdoored model generates the target sequence with abnormally high and consistent confidence compared to benign generation. Building on this insight, we propose ConfGuard, a lightweight and effective detection method that monitors a sliding window of token confidences to identify sequence lock. Extensive experiments demonstrate ConfGuard achieves a near 100% true positive rate (TPR) and a negligible false positive rate (FPR) in the vast majority of cases. Crucially, the ConfGuard enables real-time detection almost without additional latency, making it a practical backdoor defense for real-world LLM deployments.

In the digital age of the Internet of Things, users rely extensively on online accounts to access a variety of IoT services and applications. However, password leakage significantly threatens users’ privacy, assets, and reputations, making passwords primary targets for cyberattacks, including credential-guessing attacks. To address this vulnerability, this paper proposes FastPaD, a Fast privacy-preserving protocol designed to detect password similarity leakage using homomorphic encryption. FastPaD employs a novelhorizontal homomorphic batch encoding method, facilitating efficient batch detection of similar passwords. The protocol also incorporates optimization strategies such aspolynomial partitioning and power computation windowing, significantly reducing the depth of homomorphic multiplications. This enables the use of smaller encryption parameters, which improves computational efficiency. Moreover, FastPaD features a customizable mechanism to balance functionality and performance, effectively optimizing the trade-off between communication and computational overhead. Experimental results demonstrate that in scenarios without similarity detection, FastPaD achieves a detection computation time of 0.17 seconds and a total communication overhead of 3.88 MB, surpassing state-of-the-art protocols. When similarity detection is enabled, FastPaD completes the detection process in 6.45 seconds with a communication overhead of 10.64 MB. Compared to the Pipa protocol, FastPaD provides approximately a 2.70× improvement in computation time and a 22.27× improvement in communication efficiency. Relative to Yu’s protocol, FastPaD demonstrates a 15.78× faster computation and a 1.31× reduction in communication overhead.

Box-free model watermarking has gained significant attention in deep neural network (DNN) intellectual property protection due to its model-agnostic nature and its ability to flexibly manage high-entropy image outputs from generative models. Typically operating in a black-box manner, it employs an encoder-decoder framework for watermark embedding and extraction. While existing research has focused primarily on the encoders for the robustness to resist various attacks, the decoders have been largely overlooked, leading to attacks against the watermark. In this paper, we identify one such attack against the decoder, where query responses are utilized to obtain backpropagated gradients to train a watermark remover. To address this issue, we propose Decoder Gradient Shields (DGSs), a family of defense mechanisms, including DGS at the output (DGS-O), at the input (DGS-I), and in the layers (DGS-L) of the decoder, with a closed-form solution for DGS-O and provable performance for all DGS. Leveraging the joint design of reorienting and rescaling of the gradients from watermark channel gradient leaking queries, the proposed DGSs effectively prevent the watermark remover from achieving training convergence to the desired low-loss value, while preserving image quality of the decoder output. We demonstrate the effectiveness of our proposed DGSs in diverse application scenarios. Our experimental results on deraining and image generation tasks with the state-of-the-art box-free watermarking show that our DGSs achieve a defense success rate of 100% under all settings.

Deep learning technologies have seen widespread adoption in WiFi-based wireless sensing systems. However, they are inherently vulnerable to adversarial perturbation attacks, which has received little attention within the WiFi sensing community. To more comprehensively understand the potential threats posed by perturbation attacks, we present a novel attack method, named WiIntruder, distinguishing itself with universality, robustness, and stealthiness. This paper intends to provide a catalyst that promotes the assessment of security in existing WiFi-based sensing systems. We achieve the three aforementioned salient features in WiIntruder through the following three steps: (1) Maximizing transferability by differentiating user-state-specific feature spaces across sensing models, thereby enabling a universal perturbation attack vector applicable to a wide range of applications; (2) Mitigating the impact of perturbation signal distortion by optimizing key factors of device synchronization and wireless propagation through a heuristic particle swarm algorithm; and (3) Enhancing the diversity and stealthiness of attack patterns by randomly switching among perturbation surrogates generated by a generative adversarial network. Experimental results confirm the threat posed by WiIntruder to four common WiFi-based services, with the average accuracy decrease by 72.9% under black-box attack scenarios.

Federated learning (FL) is a paradigm that ensures the confidentiality and accessibility of data without requiring the collection of private data from multiple sources. It acquires an aggregation model by integrating various local models from clients. However, clients are vulnerable to numerous security and privacy threats. Existing solutions were unable to implement training models that are both dropout-resilient and lightweight while also providing verification capabilities when large-scale clients are involved in federated training. To improve the usability of FL, we propose a verifiable and lightweight multi-round secure FL framework by designing and incorporating a double-masking mechanism to ensure secure transmission. Moreover, we optimize the secure aggregation strategy by designing a dropout-resilience method via the secret-sharing mechanism. Specifically, we establish a lightweight model-secure training scheme and provide a parameter reuse strategy by constructing a full connection graph, which reduces computational cost and communication overhead. Furthermore, we propose a secure authentication protocol that enables the client to verify the accuracy of the computing results from the server. Extensive experimental evaluations indicate that our solution demonstrates relatively modest performance but superior functionality compared to current state-of-the-art methods. In particular, we can achieve the verification function with an acceptable increase in computational cost of approximately 200ms per epoch.

Decentralized finance has experienced phenomenal growth, revolutionizing the landscape of financial transactions and asset management via blockchain. Yet, this swift growth brings with it substantial challenges, notably the surge in scam tokens, imposing significant security threats on cryptocurrency investments and trading. Existing detection methods of scam token, primarily relying on analyzing contract codes or transaction patterns, struggle to catch increasingly sophisticated tactics employed by scammers. For example, contract-based analysis are unable to identify scams lacking overt malicious code, e.g., most rugpulls, while transaction-based methods generally lack the foresight to early-detect potential risks. In this paper, we present TOKENSCOUT, the first temporal GNN-based framework for scam token early detection. TOKEN SCOUT formulates token transfer data as a dynamic temporal attributed multigraph and leverages the temporal graph learning model to learn graph representations. It also builds a graph rep resentation refining model based on contrastive learning to learn a more discriminative representation space for risk identification. We evaluated TOKENSCOUT using a comprehensive dataset of 214,084 standard ERC20 tokens from 2015 to February 2023. TOKENSCOUT achieves a balanced accuracy of 98.41%. Additionally, from March to May 2023, deploying TOKENSCOUT on Ethereum effectively identified 706 rugpulls, 174 honeypots, and 90 Ponzi schemes, thereby alerting to potential risks exceeding $240 million.

The combination of Deep Learning (DL) and Federated Learning (FL) makes it a popular paradigm to train powerful models securely on large-scale data in a distributed way. However, current solutions face challenges such as significant communication overheads for clients with limited resources, potential privacy risks arising from FL's distributed nature, and the inability to maintain model accuracy without loss under high compression ratios. To solve these issues, we propose a lightweight Communication-efficient and Privacy-preserving FL scheme CPFL by designing Cyclic Segmented Compressive Sensing (CSCS) and using efficient Symmetric Homomorphic Encryption (SHE), which greatly reduces the number of transmitted model weights without sacrificing model accuracy. Formal analysis shows the security of CPFL against known-plaintext attacks and ensures model convergence. Extensive experiments demonstrate that CPFL achieves remarkable model accuracy under more than 200× compression ratio, and even reduces the communication cost by 99.5% compared with previous solutions.

Robot vacuum cleaners (RVCs) with autonomous navigation and decision-making capabilities have become an integral part of modern homes. During their operations, these devices may inadvertently enter privacy-sensitive areas, leading to potential privacy breaches. However, existing defense methods risk exposing the location of private areas, require root privileges, or are designed for infrared sensors that are ineffective for camera-based RVCs. To overcome these limitations, we propose a novel solution, a ground-view adversarial patch named GPatch, preventing RVCs from entering privacy-sensitive areas. Users only need to place GPatch at the entrance of restricted areas to prevent an RVC's unauthorized access, while also providing a warning to unauthorized individuals. We evaluate GPatch in realworld environments with an average success rate of 87.27%, and experimental results demonstrate its effectiveness, robustness, and transferability, making it a practical, user-friendly, and reliable solution for safeguarding privacy in home environments.

Traffic sign recognition (TSR) systems are crucial for autonomous driving but are vulnerable to backdoor attacks. Existing physical backdoor attacks either lack stealth, provide inflexible attack control, or ignore emerging Vision-Large-Language-Models (VLMs). In this paper, we introduce FIGhost, the first physical-world backdoor attack leveraging fluorescent ink as triggers. Fluorescent triggers are invisible under normal conditions and activated stealthily by ultraviolet light, providing superior stealthiness, flexibility, and untraceability. Inspired by real-world graffiti, we derive realistic trigger shapes and enhance their robustness via an interpolation-based fluorescence simulation algorithm. Furthermore, we develop an automated backdoor sample generation method to support three attack objectives. Extensive evaluations in the physical world demonstrate FIGhost's effectiveness against state-of-the-art detectors and VLMs, maintaining robustness under environmental variations and effectively evading existing defenses.

Pre-trained language models (PTLMs) have become integral to modern natural language processing (NLP), yet their reuse exposes them to supply chain risks such as backdoor attacks. Existing studies assume that attackers target specific downstream tasks, overlooking how a backdoored PTLM behaves when fine-tuned for unrelated applications. In practice, such unintended adaptation can trigger anomalous and inconsistent predictions, revealing the backdoor and compromising its stealthiness. We define this phenomenon asbackdoor complications, i.e., unintended behavioral side effects emerging on non-target tasks. This work presents the first systematic quantification and mitigation of backdoor complications. Through extensive experiments on 3 widely used PTLMs and 15 benchmark datasets, we show that complications are pervasive across both single- and multi-task attack settings, causing triggered outputs to collapse into arbitrary classes. To address this issue, we propose theComplication-Suppressed Backdoor Attack(CSBA), a task-agnostic, multi-objective framework that leverages auxiliary non-target datasets to suppress backdoor complications. CSBA effectively suppresses complications on unseen downstream tasks while maintaining near-perfect attack success rates. Our work reveals a critical side effect in backdoored PTLMs and provides a new perspective on the stealthiness and robustness of model supply chain security.

Vision-Language Models (VLMs) are increasingly deployed in real-world applications, but their high inference cost makes them vulnerable to resource consumption attacks. Prior attacks attempt to extend VLM output sequences by optimizing adversarial images, thereby increasing inference costs. However, these extended outputs often introduce irrelevant abnormal content, compromising attack stealthiness. This trade-off between effectiveness and stealthiness poses a major limitation for existing attacks. To address this challenge, we proposeHidden Tail, a stealthy resource consumption attack that crafts prompt-agnostic adversarial images, inducing VLMs to generate maximum-length outputs by appending special tokens invisible to users. Our method employs a composite loss function that balances semantic preservation, repetitive special token induction, and suppression of the end-of-sequence (EOS) token, optimized via a dynamic weighting strategy. Extensive experiments show thatHidden Tailoutperforms existing attacks, increasing output length by up to 19.2× and reaching the maximum token limit, while preserving attack stealthiness. These results highlight the urgent need to improve the robustness of VLMs against efficiency-oriented adversarial threats. Our code is available athttps://github.com/zhangrui4041/Hidden_Tail.

Recently, state-of-the-art stereo-matching networks trained on large-scale synthetic data have shown remarkable performance. However, their capacity to extrapolate effectively to unseen real-world data,i.e.different domains, remains a challenge. The major difficulty resides in the unforeseeable domain gap when generalizing from synthetic data to real-world data. In this paper, we introduceADASM, an approach using adversarial domain alignment, designed to enhance the robustness and generalization of stereo-matching networks. It mainly consists of two modules: an end-to-end robustness optimizer and a domain-invariant feature learner. First, we adapt adversarial training into the stereo-matching task to reduce models' sensitivity to the perturbation in real-world samples. By introducing worst cases into the training space, we take unseen data into account and achieve robust disparity estimation for the end-to-end model. Then, via simulating the real-world noise with gradient-based perturbation, we construct a fictitious domain, which is taken as a referential distribution of the real-world noisy data, for further domain alignment. Specifically, we propose to utilize Maximum Mean Discrepancy to realize domain regularization between the original domain and the fictitious one. Finally, we fuse all aforementioned objectives and propose a unified, simple but effective loss function that can be adapted toallstereo-matching networks. The extensive experiments show that our method achieves a superior disparity estimation performance on various real-world benchmarks, including KITTI, Middlebury, and DrivingStereo. More importantly,ADASMobtains competitive or even better performance than the fine-tuning strategy, revealing its fine-tuning-free character.

Third-party applications (TPAs) are integral components of managed Kubernetes clusters, but are also frequently exploited in takeover attacks. Recent incidents have demonstrated that TPAs can be weaponized to gain control over clusters. Given their critical role within the Kubernetes ecosystem, it is essential to explore the potential attack surfaces associated with various types of TPAs. To address this, we propose${\sf KubeSec}$, a framework that systematically investigates these risks by analyzing application permission configurations and component code dependencies. This investigation revealed a significant number of insecure RBAC binding patterns, uncovering 562 such patterns and identifying 375 vulnerabilities linked to 134 CVEs. These vulnerabilities impact millions of users, with an average remediation time exceeding 10 months. All findings have been reported to the relevant teams, leading to the assignment of 21 new CVEs by the community. These results highlight substantial security risks associated with TPAs in Kubernetes clusters and emphasize the urgent need for further research to develop more secure cluster management practices.

Recently, private and robust federated learning (FL) schemes have been proposed to address privacy inference and Byzantine attacks simultaneously. However, existing schemes are inefficient in private and robust aggregation protocols due to the employment of heavy cryptographic techniques. To approach the above problem, we propose Sanitizer, an efficient, private, and robust FL framework. Specifically, we first design a Byzantine-robust defense for communication-efficient sign-based FL. We further propose a customized private and robust aggregation scheme built on our Byzantine-robust defense for FL. The core of our construction is two new efficient protocols, i.e.,high-dimensional boolean summationandweighted boolean majority vote, which serve as the main building blocks of Sanitizer. Extensive evaluations on real-world datasets demonstrate that Sanitizer is blazing fast, achieving 19 ∼ 23× less runtime compared to the state-of-the-art. Meanwhile, Sanitizer achieves the same accuracy as the plaintext and superior Byzantine robustness against various classic attacks.

In the Machine Learning as a Service paradigm, a service provider (e.g., a server) hosting a model offers inference APIs to clients, who can send their queries and receive the inference results. While most recent secure inference works focus on addressing privacy issues, they overlook the importance of checking the service quality and reliability. A malicious server may deviate from the protocol specification to deliberately provide incorrect services such as using low-quality models. Thus, it is necessary to design new solutions to empower clients to verify the server’s model accuracy and inference integrity while protecting both parties’ privacy. We present Conan, a new secure and reliable inference framework against malicious servers to achieve accuracy verification, inference integrity, and privacy simultaneously. In Conan, the server first commits to the model and proves in zero-knowledge that the committed model achieves the claimed accuracy. Then both parties perform secure inference on the committed model against the malicious server. To instantiate the above framework, we design generic maliciously secure two-party computation (2PC) protocols with a fixed corrupted party, which may be of independent interest. Our protocols achieve high efficiency by utilizing the advantage that the semi-honest party can check the behavior of the corrupted party. Furthermore, they support both arithmetic and Boolean circuit evaluation, a crucial attribute for secure inference on complicated machine learning models. We implement the fixed-corruption 2PC protocols for our secure and reliable inference. The experimental results show 1 ~ 2 orders of magnitude improvements over conventional maliciously secure protocols in terms of communication and computation costs.

Privacy labels (e.g., Data Safety section on Google Play) aim to replace lengthy privacy policies with concise and standardized summaries of in-app privacy practices. However, studies reveal widespread inaccuracies in these self-reported labels, with developers omitting or misrepresenting privacy practices, undermining user trust and regulatory compliance. Existing methods for detecting such discrepancies lack coverage or scalability and fail to address the semantic ambiguity inherent in privacy label auditing. We present Iterative Context Reconstruction (ICR), an evidence-driven workflow that reconstructs context from decompiled code to resolve the ambiguity. Based on ICR, PriLabel: Towards Comprehensively Uncovering Omitted Disclosures in Privacy Labels of Android Apps on a Large Scale is a context-aware static auditor that comprehensively uncoversomitted disclosuresin Android privacy labels, mapping transmitted data to Google’s label taxonomy in asource-freeandontology-freemanner. Our evaluation demonstrates PriLabel: Towards Comprehensively Uncovering Omitted Disclosures in Privacy Labels of Android Apps on a Large Scale’s high precision (91.5%) in detecting omitted disclosures in privacy labels. Applied to 4,851 top-installed Google Play apps, it revealed that 2,374 apps omitted at least one disclosure, with 210 transmitting sensitive financial data (e.g., credit card numbers) without proper labeling, exposing systemic risks of non-compliance.

Collaborative Perception (CP) has been shown to be a promising technique for multi-agent autonomous driving and multi-agent robotic systems, where multiple agents share their perception information to enhance the overall perception performance and expand the perception range. However, in CP, an ego agent needs to receive messages from its collaborators, which makes it vulnerable to attacks from malicious agents. To address this critical issue, we propose a unified, probability-agnostic, and adaptive framework, namely, CP-uniGuard, which is a tailored defense mechanism for CP deployed by each agent to accurately detect and eliminate malicious agents in its collaboration network. Our key idea is to enable CP to reach a consensus rather than a conflict against an ego agent's perception results. Based on this idea, we first develop a probability-agnostic sample consensus (PASAC) method to effectively sample a subset of the collaborators and verify the consensus without prior probabilities of malicious agents. Furthermore, we define collaborative consistency loss (CCLoss) for object detection task and bird's eye view (BEV) segmentation task to capture the discrepancy between an ego agent and its collaborators, which is used as a verification criterion for consensus. In addition, we propose online adaptive threshold via dual sliding windows to dynamically adjust the threshold for consensus verification and ensure the reliability of the systems in dynamic environments. Finally, we conduct extensive experiments and demonstrate the effectiveness of our framework.

Collaborative Perception (CP) has shown a promising technique for autonomous driving, where multiple connected and autonomous vehicles (CAVs) share their perception information to enhance the overall perception performance and expand the perception range. However, in CP, ego CAV needs to receive messages from the collaborators, which makes it easy to be attacked by malicious agents. For example, a malicious agent can send harmful information to the ego CAV to mislead it. To address this critical issue, we propose a novel method, **CP-Guard**, a tailored defense mechanism for CP that can be deployed by each agent to accurately detect and eliminate malicious agents in its collaboration network. Our key idea is that CP will lead to a consensus rather than a conflict against the ego CAV's perception results. Based on this idea, we first develop a probability-agnostic sample consensus (PASAC) method that can effectively sample a subset of the collaborators and verify the consensus without prior probabilities of malicious agents. Furthermore, we design a collaborative consistency loss (CCLoss) to calculate the discrepancy between the ego CAV and the collaborators, which is used as a verification criterion for consensus. Finally, we conduct extensive experiments in collaborative bird's eye view (BEV) tasks and the results demonstrate the effectiveness of our CP-Guard.

Substitute training-based data-free black-box attacks pose a significant threat to enterprise-deployed models. These attacks use a generator to synthesize data and query APIs, then train a substitute model to approximate the target model's decision boundary based on the returned results. However, existing attack methods often struggle to produce sufficiently diverse data, particularly for complex target models and extensive target data domains, severely limiting their practical application. To address this gap, we design domain-augmented learning to improve the quality of the synthetic data domain (SDD) generated by the generator from two perspectives. Specifically, (1) To broaden the SDD's coverage, we introduce textual semantic embeddings into the generator for the first time. (2) For enhancing the SDD's discretization, we propose a competitive optimization strategy that forces the generator to self-compete, along with heterogeneity excitation to overcome the constraints of information entropy on diversity. Comprehensive experiments demonstrate that our method is more effective. In non-targeted attacks on the CIFAR-10 and Tiny-ImageNet datasets, our method outperforms the state-of-the-art by 14% and 7% in attack success rate, respectively.

Single Sign-On (SSO) allows users to sign on to multiple relying providers (RPs) with a single authentication token issued by an identity provider (IdP), which provides users a convenient and efficient way to access multiple services from different RPs. As the security of SSO relies on the reliability of IdP (which needs to well maintain a secret used to issue tokens), it suffers from the single-point-of-failure problem. Existing schemes address the problem by utilizing multiple IdPs to issue tokens in a threshold way, so as to make the task of compromising the secret for adversaries as difficult as possible. However, no security guarantee is considered once the secret is compromised by adversaries. In this paper, we propose a distributed forward-secure SSO scheme, dubbed DFSSO, to achieve security in the “post-compromising case” with minimized costs: after the secret is compromised, only a small portion of users need to re-authenticate themselves with IdPs. The key technique behind DFSSO is a new cryptographic primitive, i.e., threshold forward-secure signature, which is interesting in its own right. We integrate DFSSO into OpenID Connect (i.e., OIDC, a popular SSO standard), implement a prototype, and conduct a comprehensive performance evaluation, which demonstrates that DFSSO is efficient and practical.

The intellectual property of deep image-to-image models can be protected by the so-called box-free watermarking. It uses an encoder and a decoder, respectively, to embed into and extract from the model’s output images invisible copyright marks. Prior works have improved watermark robustness, focusing on the design of better watermark encoders. In this paper, we reveal an overlooked vulnerability of the unprotected watermark decoder which is jointly trained with the encoder and can be exploited to train a watermark removal network. To defend against such an attack, we propose the decoder gradient shield (DGS) as a protection layer in the decoder API to prevent gradient-based watermark removal with a closed-form solution. The fundamental idea is inspired by the classical adversarial attack, but is utilized for the first time as a defensive mechanism in the box-free model watermarking. We then demonstrate that DGS can reorient and rescale the gradient directions of watermarked queries and stop the watermark remover’s training loss from converging to the level without DGS, while retaining decoder output image quality. Experimental results verify the effectiveness of the proposed method. Code of paper is available at https://github.com/haonanAN309/CVPR-2025-Official-Implementation-Decoder-Gradient-Shield.

Deep learning models employed in face recognition (FR) systems have been shown to be vulnerable to physical adversarial attacks through various modalities, including patches, projections, and infrared radiation. However, existing adversarial examples targeting FR systems often suffer from issues such as conspicuousness, limited effectiveness, and insufficient robustness. To address these challenges, we propose a novel approach for adversarial face generation, UVHat, which utilizes ultraviolet (UV) emitters mounted on a hat to enable invisible and potent attacks in black-box settings. Specifically, UVHat simulates UV light sources via video interpolation and models the positions of these light sources on a curved surface, specifically the human head in our study. To optimize attack performance, UVHat integrates a reinforcement learning-based optimization strategy, which explores a vast parameter search space, encompassing factors such as shooting distance, power, and wavelength. Extensive experimental evaluations validate that UVHat substantially improves the attack success rate in black-box settings, enabling adversarial attacks from multiple angles with enhanced robustness.

Recent research highlights concerns about the trustworthiness of third-party Pre-Trained Language Models (PTLMs) due to potential backdoor attacks. These backdoored PTLMs, however, are effective only for specific pre-defined downstream tasks. In reality, these PTLMs can be adapted to many other unrelated downstream tasks. Such adaptation may lead to unforeseen consequences in downstream model outputs, consequently raising user suspicion and compromising attack stealthiness. We refer to this phenomenon as backdoor complications. In this paper, we undertake the first comprehensive quantification of backdoor complications. Through extensive experiments using 4 prominent PTLMs and 16 text classification benchmark datasets, we demonstrate the widespread presence of backdoor complications in downstream models fine-tuned from backdoored PTLMs. The output distribution of triggered samples significantly deviates from that of clean samples. Consequently, we propose a backdoor complication reduction method leveraging multi-task learning to mitigate complications without prior knowledge of downstream tasks. The experimental results demonstrate that our proposed method can effectively reduce complications while maintaining the efficacy and consistency of backdoor attacks.

Text-to-image diffusion models have advanced controllable image generation, with ControlNet plugins enabling precise structural guidance and domain-specific adaptations. As these plugins become widely shared and personalized, protecting their ownership and preventing misuse becomes crucial. Existing watermarking methods address robustness against fine-tuning and personalization at the model level, but fail to address ControlNet-like plugins or modules specifically. To address this gap, we propose CtrlMark, the first watermarking framework designed specifically for ControlNet plugin modules. CtrlMark embeds a robust, triggerable watermark as a benign backdoor, activated by a composite trigger combining text and structural inputs. Furthermore, CtrlMark achieves few misactivations and strong robustness against downstream fine-tuning, by watermark penalization and leveraging a fixed latent residual embedding localized to a spatial region. Extensive experiments demonstrate CtrlMark maintains high watermark activation rates and visual fidelity, providing effective and practical protection for modular ControlNet components in diverse generation scenarios.

Natural language processing (NLP) systems have been demonstrated to be vulnerable to backdoor attacks. Specifically, attackers embed the backdoor into the model by poisoning training data, producing the desired results when the input contains pre-defined triggers. Typical textual backdoor attacks adopt static triggers such as words or phrases, which make them detectable by existing defense methods. To enhance stealthiness, this paper introduces a hidden backdoor attack method utilizing formal text style transfer (FTST). Specifically, we adopt a formal text style transfer model to convert part of the benign training samples into formal samples, which serve as the backdoor samples. Compared to static textual triggers, FTST-based triggers can maintain original semantics while evading common defenses and human detections. We conduct extensive experiments on typical NLP tasks, including topic and sentiment classification tasks utilizing three prominent pre-trained language models and four datasets. The results show that our approach achieves the desired attack performance while preserving the normal-functionality of the model. Furthermore, compared to common word-level triggers and sentence-level triggers, our approach has been demonstrated to be more stealthy under GPT-2-based perplexity detection and more robust under backdoor defense methods.

Recently, traffic sign recognition (TSR) systems have become a prominent target for physical adversarial attacks. These attacks typically rely on conspicuous stickers and projections, or using invisible light and acoustic signals that can be easily blocked. In this paper, we introduce a novel attack medium, i.e., fluorescent ink, to design a stealthy and effective physical adversarial patch, namely FIPatch, to advance the state-of-the-art. Specifically, we first model the fluorescence effect in the digital domain to identify the optimal attack settings, which guide the real-world fluorescence parameters. By applying a carefully designed fluorescence perturbation to the target sign, the attacker can later trigger a fluorescent effect using invisible ultraviolet light, causing the TSR system to misclassify the sign and potentially leading to traffic accidents. We conducted a comprehensive evaluation to investigate the effectiveness of FIPatch, which shows a success rate of 98.31% in low-light conditions. Furthermore, our attack successfully bypasses five popular defenses and achieves a success rate of 96.72%.

Abstract Artificial intelligence (AI) is revolutionizing both industries and reshaping the global economy. However, the rapid advancement of AI technologies brings significant security and privacy challenges. Recent incidents highlight vulnerabilities in AI systems, such as data leakage and malicious code injection, leading to severe financial losses and privacy breaches. Although existing studies have discussed specific security threats, they often lack detailed granularity and cover a limited scope. In this survey, we fill this gap by systematically categorizing and analyzing the threats and countermeasures in AI systems, which span both the training and inference stages, encompass centralized and distributed settings, and address both conventional and foundation AI models. By reviewing existing literature, we aim to provide AI researchers and practitioners with a thorough understanding of system vulnerabilities and current countermeasures. We hope to inspire further research into robust solutions, ultimately contributing to the development of resilient AI technologies.

With the rapid development of deep learning technology, deep learning-based Image-to-Image (I2I) networks have become the predominant choice for I2I tasks like image super-resolution and denoising. Despite their remarkable performance, the security of I2I networks has not been thoroughly investigated. While some studies have probed their susceptibility to adversarial attacks, none have explored the backdoor attack against I2I networks, which is a more stealthy and severe threat. In this work, for the first time, we comprehensively investigate the vulnerability of I2I networks to backdoor attacks. We propose a backdoor attack against I2I tasks, where the backdoored I2I network behaves normally on clean input images, yet outputs a specific inappropriate image when the backdoor trigger appears on the input image. To achieve such an I2I backdoor attack, we design a universal adversarial perturbation (UAP) generation algorithm for I2I networks, where the generated UAP is used as the trigger for the I2I backdoor. Besides, multi-task learning (MTL) with dynamic weighting methods is employed in the backdoor training process to gain better results. Expanding our focus beyond I2I tasks, we extend our I2I backdoor to attack downstream tasks, including image classification and object detection. Specifically, the backdoor-triggered image processed by the backdoored image denoising network can fool the downstream image classifiers and object detectors. Extensive experiments demonstrate the effectiveness of the I2I backdoor on state-of-the-art I2I network architectures as well as the robustness against different backdoor defenses.

Deep neural networks (DNNs) are known to be susceptible to various malicious attacks, such as adversarial and backdoor attacks. However, most of these attacks utilize additive adversarial perturbations (or backdoor triggers) within an$L_{p}$-norm constraint. They can be easily defeated by image preprocessing strategies, such as image compression and image super-resolution. To address this limitation, instead of using additive adversarial perturbations (or backdoor triggers) in the pixel space, this work revisits the design of adversarial perturbations (or backdoor triggers) from the perspective of color space and conducts a comprehensive analysis. Specifically, we propose a color space backdoor attack and a color space adversarial attack where the color space shift is used as the trigger and perturbation. To find the optimal trigger or perturbation in the black-box scenario, we perform an iterative optimization process with the Particle Swarm Optimization algorithm. Experimental results confirm the robustness of the proposed color space attacks against image preprocessing defenses as well as other mainstream defense methods. In addition, we also design adaptive defense strategies and evaluate their effectiveness against color space attacks. Our work emphasizes the importance of the color space when developing malicious attacks against DNN and urges more research in this area.

In this paper, we present$\mathtt {Antelope}$, a semi-honest large-scale secure inference system without revealing either clients’ data or model parameters. The main contributions of$\mathtt {Antelope}$are new two-party computation (2PC) protocols over a ring$\mathbb {Z}_{2^\ell }$for non-linear layers, which optimize the online computation and communication overhead thus outperforming the state-of-the-art 2PC systems. Specifically, we reformulate the comparison function as an Equality-to-Zero test followed by multiplication, decoupling the bit-wise rounding dependency in traditional secret sharing-based bit extraction. With this technique, the evaluation of the ReLU non-linear activation function is$1.7\times$-$84.5\times$faster than existing solutions in online communication cost. We also develop a suite of optimizations that improve the efficiency of secure division protocols, which are tailored to different divisor settings in the neural networks. We extend our protocols to construct efficient implementations for several building blocks such as ReLU, Maxpool, truncation, and Softmax. End-to-end evaluation on realistic ImageNet-scale networks demonstrates that$\mathtt {Antelope}$achieves over$22.3\times$and$23.0\times$online runtime speedups in LAN and WAN settings, respectively, without accuracy loss, compared to the state-of-the-art works.

Fueled by its successful commercialization, the recommender system (RS) has gained widespread attention. However, as the training data fed into the RS models are often highly sensitive, it ultimately leads to severe privacy concerns, especially when data are shared among different platforms. In this paper, we follow the tune of existing works to investigate the problem of secure sparse matrix multiplication for cross-platform RSs. Two fundamental and critical issues are addressed: preserving the training data privacy and breaking the data silo problem. Specifically, we propose two concrete constructions with significantly boosted efficiency. They are designed for the sparse location insensitive case and location sensitive case, respectively. State-of-the-art cryptography building blocks including homomorphic encryption (HE) and private information retrieval (PIR) are fused into our protocols with non-trivial optimizations. As a result, our schemes can enjoy the HE acceleration technique without privacy trade-offs. We give formal security proofs for the proposed schemes and conduct extensive experiments on both real and large-scale simulated datasets. Compared with state-of-the-art works, our two schemes compress the running time roughly by$10\times$and$2.8\times$. They also attain up to$15\times$and$2.3\times$communication reduction without accuracy loss.

The increasing demand for secure and efficient data sharing has underscored the importance of developing robust cryptographic schemes. However, many existing endeavors have overlooked the following critical issues: (1) unauthorized access resulting from malicious information leakage by senders; (2) absence of constraints on write and read permissions for participants; (3) and inflexibility of strategies to dynamically designate ciphertexts to multiple recipients. In this paper, we present SCPA, a cross-domain access control scheme imbued with sanitization features and propelled by policy-driven dynamic authorization, tailored for cloud-based data sharing. This scheme not only facilitates access controls, including regulations for no-read and no-write stipulations, governing the data permissible for senders to transmit and recipients to acquire but also enables the dynamic sharing of a data ciphertext subset with additional recipients beyond the originally sanctioned ones. We also provide comprehensive security proofs rigorously indicating the security of the invented SCPA. Moreover, to assess the efficacy of our SCPA, we undertake thorough theoretical and experimental analyses, showcasing its feasibility and superior performance.

Sparse inner product (SIP) has the attractive property of overhead being dominated by the intersection of inputs between parties, independent of the actual input size. It has intriguing prospects, especially for boosting machine learning on large-scale data, which are tangled with sparse data. In this paper, we investigate privacy-preserving SIP problems that have rarely been explored before. Specifically, we propose two concrete constructions, one requiring offline linear communication, which can be amortized across queries, while the other has sublinear overhead but relies on the more computationally expensive tool. Our approach exploits state-of-the-art cryptography tools including garbled Bloom filters (GBF) and Private Information Retrieval (PIR) as the cornerstone but carefully fuses them to obtain nontrivial overhead reductions. We provide formal security analysis of the proposed constructions and implement them into representative machine learning algorithms including k-nearest neighbors, naive Bayes classification, and logistic regression. Compared to existing efforts, our method achieves 2-$50\times$speedup in runtime and up to$10\times$reduction in communication.

Federated learning (FL) is a distributed machine learning technique that guarantees the privacy of user data. However, FL has been shown to be vulnerable to gradient leakage attacks (GLA), which have the ability to reconstruct private training data from public gradients with high probability. These attacks are either analytic-based, requiring modification of the FL model, or optimization-based, requiring long convergence times and failing to effectively address the challenge of dealing with highly compressed gradients in practical FL systems. This paper presents a pioneering generation-based GLA method called FGLA that can reconstruct batches of user data without the need for the optimization process. We specifically design a feature separation technique that first extracts the features of each sample in a batch and then directly generates the user data. Our extensive experiments on multiple image datasets show that FGLA can reconstruct user images in seconds with a batch size of 256 from highly compressed gradients (0.8% compression ratio or higher), thereby significantly outperforming state-of-the-art methods.

Ample profits of GPU cryptojacking attract hackers to recklessly invade victims’ devices, for completing specific cryptocurrency mining tasks. Such malicious invasion undoubtedly obstructs normal device usage and wastes computation resources. To resist the threat of GPU cryptojacking, existing works aim to timely detect and clear away it, by distinguishing the dissimilitude between it and legitimate applications. However, these detection mechanisms inappropriately rely on two conflict cornerstones, manifested in leveragingmutable samples of illegitimate cryptojackingto designsupervision-based detection models requiring samples with stable patterns. This limitation compromises the practicability of existing detection mechanisms in the face of mutable cryptojacking samples. To fill the gap, we explore the superiority of unsupervised learning in handling this issue and further propose an unsupervised manner-enabled detection mechanism named MagInspector, only using legitimate applications’ magnetic signatures from GPU side channels for model construction. MagInspector innovates in training an unsupervised autoencoder network by an adversarial mode that well learns the stable signature patterns of legitimate applications, while incompatible with mutable cryptojacking ones. In the process of model training, we elaborately extract mutual energy cumulation distribution features to represent legitimate applications to overcome the impact of their inter-type differences. Meanwhile, a locality sensitive hashing-driven outlier removal algorithm is designed to enhance MagInspector’s robustness to the noise samples. Finally, extensive experiments are conducted on GPUs covering four generations of common NVIDIA architectures and two generations of AMD architectures; the results show that applying MagInspector to mutable cryptojacking signature detection achieves a significant average accuracy improvement of 25.5% and 17.8%, respectively.

Using side-effect sensing information to monitor the behavior of smartphone usage raises privacy leakage concerns. However, existing research typically utilizes only a single sensing channel or performs a simple aggregation of multi-channel data to infer user behavior, without sufficiently leveraging rich spatiotemporal information embedded in the diverse sensing channels. Such a narrow focus of existing works fails to exhibit the real risk of user privacy leakage. To bridge this research gap, we propose HiddenSpy, a comprehensive study assessing the smartphone usage snooping associated with multiple sensing channels, such as accelerometers and magnetometers. We start by examining the relationship between the data gathered from each channel and daily usage behaviors, highlighting information volume differences across channels. Building on this analysis, we propose a multi-layer attention mechanism that dynamically adjusts the importance of spatiotemporal information from different channels and time frames, facilitating the efficient use of multi-channel data for behavior inference. Importantly, our work marks a pivotal shift from addressing information leakage in single channels to managing information exposure throughout the smartphone sensing system, laying the foundation for more comprehensive protective measures. To validate our approach, we collect data from forty widely-used applications and evaluate the corresponding usage behavior snooping performance. The results show that HiddenSpy improves accuracy in three common snooping tasks, while its defense mechanism reduces accuracy to a low level, effectively preventing information leakage.

Privacy concerns have been persistently afflicting individuals within online social networks (OSNs), rendering privacy-preserving communications over the Internet with authentication especially important. Unfortunately, the guarantees of privacy and authenticity are not always provided in OSNs. Individuals are still facing the challenges of being deceived or exploited. To mitigate these issues, anonymous social networks (ASNs) have emerged as a remedy for OSNs, facilitating individuals to connect with others anonymously and authentically. Despite the existence of numerous and remarkable cryptographic primitives, there are no formal solutions for ASNs except for matchmaking encryption (ME), since ME can simultaneously provide various key functionalities, i.e. bilateral access control, identity anonymity, and message authentication, to address the requirements of ASNs. In this paper, we design a system for ASNs by adopting fuzzy identity-based matchmaking encryption (fuzzy IB-ME), and the proposed scheme in this work is highly efficient. The scheme also realizes adaptive security in generic group model (GGM), which is generally adopted in pairing-based cryptography. The proposed ASNs system offers various advantages compared to the previous solutions, including 1) bilateral access control, 2) enhanced security, 3) high scalability, and 4) high efficiency. In addition to theoretical evaluations, we conduct extensive experiments to evaluate our scheme’s computational and storage efficiency. These evaluations indicate that our solution outperforms previous solutions and as well as preserves many desired functionalities.

In this paper, we introduce SeiFS, a Secure and Lightweight Feature Selection system designed to ensure high-quality inputs for Machine Learning (ML) tasks. Unlike previous approaches involving multiple non-colluding servers, SeiFS operates in a natural ML scenario where multiple entities interact with a single server, without relying on additional strong assumptions. Our work presents intrinsic optimizations in feature selection that yield substantial performance improvements, including a customized data encoding method, a size-optimized comparison circuit, and a shared oblivious dimensionality reduction technique. The customized data encoding method, combined with an optimized secure data access protocol, reduces expensive comparison operations from$O(m)$to$O(\log m)$, where m represents the number of samples. The size-optimized comparison circuit achieves up to a quadruple reduction in size compared to naïve implementations. Additionally, the shared oblivious dimensionality reduction technique incorporates a novel approximated top-k selection algorithm, resulting in a circuit size reduction of approximately$k\times $. Comprehensive experiments conducted across various network settings demonstrate that our protocols outperform existing solutions, delivering efficiency improvements of an order of magnitude. Specifically, the end-to-end execution of SeiFS on real-life datasets achieves at least$62.7\times $improvements in runtime compared to the naïve implementation and takes up to$112.9\times $fewer runtimes than the state-of-the-art in the LAN setting.

Aiming to provide people with great convenience and comfort, smart home systems have been deployed in thousands of homes. In this paper, we focus on handling the security and privacy issues in such a promising system by customizing a new cryptographic primitive to provide the following security guarantees: 1) fine-grained, privacy-preserving authorization for smart home users and integrity protection of communication contents; 2) flexible self-sovereign permission delegation; 3) forward security of previous messages. To our knowledge, no previous system has been designed to consider these three security and privacy requirements simultaneously. To tackle these challenges, we put forward the first-ever efficient cryptographic primitive called the Forward-secure Hierarchical Delegable Signature (FS-HDS) scheme for smart homes. Specifically, we first propose a new primitive, efficient Hierarchical Delegable Signature (HDS) scheme, which is capable of supporting partial delegation capability while realizing privacy-preserving authorization and integrity guarantee. Then, we present an FS-HDS for smart homes with the efficient HDS as the underlying building block, which not only inherits all the desirable features of HDS but also ensures that the past content integrity is not affected even if the current secret key is compromised. We provide comprehensively strict security proofs to prove the security of our proposed solutions. Its performance is also validated via experimental simulations to showcase its practicability and effectiveness.

The advent of decentralized finance has ushered in a transformative era in the financial sector, leveraging blockchain technology to facilitate peer-to-peer transactions without traditional intermediaries. Amidst this innovation, the DeFi landscape faces the pervasive threat of rugpulls, where developers abruptly abandon projects post-fundraising, leaving investors with devalued assets. This growing concern highlights a critical research gap in the proactive detection and prevention of such fraudulent schemes. To combat this, we propose RUGSCREENER, a temporal graph neural network-based solution to identify rugpull risks within DeFi transactions. It employs a dynamic representation of blockchain interactions, enriched with comprehensive node attributes and effective temporal graph learning techniques based on memory and attention mechanisms, effectively capturing the rapid-moving and complex transaction patterns indicative of potential fraud. Our evaluation is based on a newly compiled Ethereum dataset that includes two subsets: an unlabeled set with 1,882,114 transactions from 29,595 tokens for temporal graph representation learning, and a labeled set with 128,819 transactions from 1,000 tokens (500 rugpull and 500 benign) for downstream evaluation. Using this dataset, RUGSCREENER achieves a balanced accuracy of 95.7% in detecting rugpull tokens. Our extensive evaluation, utilizing the Ethereum dataset comprising 1000 tokens, showcases its robust performance with a balanced accuracy of 95.7% in detecting rugpull tokens. Remarkably, RUGSCREENER surpasses existing state-of-the-art graph learning models in detecting rugpull tokens with enhanced accuracy and reliability.

Interoperability is a fundamental challenge for longenvisioned blockchain applications. A mainstream approach is using Trusted Execution Environment (TEE) to support interoperable off-chain execution. However, this incurs multiple TEE configured with non-trivial storage capabilities running on fragile concurrent processing environments, rendering current strategies based on TEE far from being practical. This paper aims to fill this gap and design a practical interoperability mechanism with simplified TEE as the underlying architecture. Specifically, we present IvyCross, a TEE-based framework that achieves lowcost, privacy-preserving, and race-free blockchain interoperability. IvyCross allows running arbitrary smart contracts across heterogeneous blockchains atop two distributed TEE-powered hosts. We design an incentive scheme based on smart contracts to stimulate the honest behavior of two hosts, bypassing the requirement of the number of TEE and large memory need. We examine the conditions to guarantee the uniqueness of Nash Equilibrium via Game Theory. Furthermore, an extended optimistic concurrency control protocol is designed to ensure the correctness of concurrent contracts execution. We formally prove the security of IvyCross in the Universal Composability (UC) framework and implement a prototype atop Bitcoin, Ethereum, and FISCO BOCS. Extensive experimental results on end-to-end performance and concurrency control demonstrate the efficiency and practicality of IvyCross.

Ultra-wideband (UWB) ranging systems are increasingly deployed in critical, security-sensitive applications due to their precise positioning and secure ranging capabilities. In this work, we introduce a practical DoS attack via reactive jamming, referred to as UWBAD+, which targets commercial UWB ranging systems by exploiting the vulnerabilities of the normalized cross-correlation process. This allows UWBAD+ to selectively and effectively disrupt ranging sessions without requiring prior knowledge of the victim devices' configurations, leading to potentially severe consequences such as property loss, unauthorized access, or vehicle theft. The enhanced effectiveness and low detectability of UWBAD+ stem from the following: (i) it can rapidly sniff the physical layer structures of unknown UWB systems, even in the presence of multiple UWB devices operating simultaneously; (ii) it blocks each ranging session efficiently by employing field-level jamming, thus exerting a significant impact on commercial UWB ranging systems; and (iii) its compact, reactive, and selective design based on COTS UWB chips, which makes it both affordable and less noticeable. We successfully executed real-world attacks on commercial UWB ranging systems produced by the three largest UWB chip vendors in the market, including Apple, NXP, and Qorvo. We disclosed our findings to Apple, relevant Original Equipment Manufacturers (OEMs), and the Automotive Security Research Group. As of the time of writing, the involved OEM has acknowledged this vulnerability in their automotive systems and has issued a${\$} 5,000$bounty as a reward.

Decentralized finance has experienced phenomenal growth, revolutionizing the landscape of financial transactions and asset management via blockchain. Yet, this swift growth brings with it substantial challenges, notably the surge in scam tokens, imposing significant security threats on cryptocurrency investments and trading. Existing detection methods of scam token, primarily relying on analyzing contract codes or transaction patterns, struggle to catch increasingly sophisticated tactics employed by scammers. For example, contract-based analysis are unable to identify scams lacking overt malicious code, e.g., most rugpulls, while transaction-based methods generally lack the foresight to early-detect potential risks.

In modern cloud platforms, it is becoming more important to preserve the privacy of guest virtual machines (VMs) from the untrusted host. To this end, Secure Encrypted Virtualization (SEV) is developed as a hardware extension to protect VMs by encrypting their memory pages and register states. Unfortunately, such confidential VMs are still vulnerable to micro-architectural side channels, and Hardware Performance Counters (HPCs) are a prominent information leakage source. To make matters worse, currently there is no systematic defense against the HPC side channels. We introduce Aegis, a unified framework for demystifying the inherent relations between the instruction execution and HPC event statistics, and defending VMs against HPC side channels with provable privacy guarantee and minimal performance overhead. Aegis consists of three modules. Application Profiler profiles the application offline and adopts information theory to quantitatively estimate the vulnerability of HPC events. Event Fuzzer leverages the fuzzing technique to automatically generate interesting inputs, i.e., instruction sequences, that can effectively alter the HPC observations. Event Obfuscator injects noisy instructions into the protected VM based on the differential privacy mechanisms for high efficiency and privacy. We present three case studies to demonstrate that Aegis can defeat different types of HPC side-channel attacks (i.e., website fingerprinting, DNN model extraction, keystroke sniffing). Evaluations show that Aegis can effectively decrease the attack accuracy from 90% to 2%, with only 3% overhead on the application execution time and 7% overhead on the CPU usage.

Generative Adversarial Networks (GANs) have been widely used in various application scenarios. Since the production of a commercial GAN requires substantial computational and human resources, the copyright protection of GANs is urgently needed. This paper presents a novel finger-printing scheme for the Intellectual Property (IP) protection of image-to-image GANs based on a trusted third party. We break through the stealthiness and robustness bottlenecks suffered by previous fingerprinting methods for classification models being naively transferred to GANs. Specifically, we innovatively construct a composite deep learning model from the target GAN and a classifier. Then we generate fingerprint samples from this composite model, and embed them in the classifier for effective ownership verification. This scheme inspires some concrete methodologies to practically protect the modern image-to-image translation GANs. Theoretical analysis proves that these methods can satisfy different security requirements necessary for IP protection. We also conduct extensive experiments to show that our solutions outperform existing strategies.

Collaborative perception among multiple connected and autonomous vehicles (CAVs) can greatly enhance perceptive capabilities by allowing vehicles to exchange supplementary information. Despite significant advances, many design challenges still remain due to channel variations and data heterogeneity among collaborative vehicles. To address these issues, we propose ACC-DA, a channel-aware collaborative perception framework to dynamically adjust the communication graph to minimize the average transmission delay while mitigating the impacts caused by data heterogeneity. More specifically, we first construct the communication graph to minimize the transmission delay according to different channel information state. We then propose an adaptive data reconstruction mechanism to dynamically adjust the rate-distortion trade-off to enhance perception efficiency while reducing the temporal redundancy during data transmissions. Finally, we conceive a domain alignment scheme to align the data distribution from different vehicles to mitigate the domain gap between different vehicles and improve the performance of the target task. Comprehensive experiments demonstrate the effectiveness of our method in comparison to the existing state-of-the-art works.

Smart cameras have gained immense popularity in commercial markets for their safety and security capabilities. Yet, the prevalent design of these intelligent camera systems often compels users to cede control of their data to poten-tially untrusted service providers, such as cloud services. This relinquishment can lead to unauthorized data access by these intermediaries, posing significant security and privacy risks. The conventional solutions have been to employ privacy-enhancing technologies to bypass these intermediaries, but at the cost of increased overhead for video streaming and sharing. In our study, we introduce SecSCS, a user-centric, blockchain-based secure camera system that incorporates essential features like video streaming, sharing, deletion, and permission restoration. SecSCS integrates a blockchain-enabled user login protocol with a secure device pairing mechanism that combines visual authorization with blockchain to flexibly manage the device ownership. We utilize blockchain to provide integrity protection for the video clips stored remotely, ensuring the video data remains tamper-proof. Furthermore, we present a video frame compression and a fast video encryption method aimed at boosting the efficiency of smart camera systems. Our evaluations show that, in comparison to the leading decentralized scheme, CaCTUs, SecSCS improves the computational and communication overhead for live streaming by a factor of 12.58 and 11.29, respectively, at a frame rate of 24 fps and a resolution of 720p.

In recent years, autonomous driving has garnered significant attention due to its potential for improving road safety through collaborative perception among connected and autonomous vehicles (CAVs). However, time-varying channel variations in vehicular transmission environments demand dynamic allocation of communication resources. Moreover, in the context of collaborative perception, it is important to recognize that not all CAVs contribute valuable data, and some CAV data even have detrimental effects on collaborative perception. In this paper, we introduce SmartCooper, an adaptive collaborative perception framework that incorporates communication optimization and a judger mechanism to facilitate CAV data fusion. Our approach begins with optimizing the connectivity of vehicles while considering communication constraints. We then train a learnable encoder to dynamically adjust the compression ratio based on the channel state information (CSI). Subsequently, we devise a judger mechanism to filter the detrimental image data reconstructed by adaptive decoders. We evaluate the effectiveness of our proposed algorithm on the OpenCOOD platform. Our results demonstrate a substantial reduction in communication costs by 23.10% compared to the non-judger scheme. Additionally, we achieve a significant improvement on the average precision of Intersection over Union (AP@IoU) by 7.15% compared with state-of-the-art schemes.

In cloud-based data marketplaces, the cardinal objective lies in facilitating interactions between data shoppers and sellers. This engagement allows shoppers to augment their internal datasets with external data, consequently leading to significant enhancements in their machine learning models. Nonetheless, given the potential diversity of data values, it becomes critical for consumers to assess the value of data before cementing any transactions. Recently, Song et al. introduced Primal (publish in ACSAC), the pioneering cloud-assisted privacy-preserving data evaluation (PPDE) strategy. This strategy relies on variants of functional encryption (FE) as the underlying framework, conferring notable performance advantages over alternative cryptographic primitives such as secure multi-party computation and homomorphic encryption. However, in this paper, we regretfully highlight that Primal is susceptible to inadvertent misuse of FE, and leaves much-desired room for performance amelioration. To combat this, we introduce a novel cryptographic primitive known as labeled function-hiding inner-product encrypted. This new primitive serves as a remedy and forms the foundation for designing the concrete framework for PPDE. Furthermore, experiments conducted on real datasets demonstrate that our framework significantly reduces the overall computation cost of the current state-of-the-art secure PPDE scheme by roughly 10× and the communication cost for the data seller by about 2×.

Deep Neural Networks (DNNs) are vulnerable to backdoor attacks, which poison the training set to alter the model prediction over samples with a specific trigger. While existing efforts mainly focus on unimodal scenarios, modern AI systems usually employ multiple modalities to improve the model performance, making multimodal backdoor attacks more practical but structurally more complex due to inherent modality interactions, multiple attack surfaces, unbalanced modality contributions, etc. These factors affect the effectiveness of backdooring multimodal learning significantly but have not been fully investigated yet.To bridge this gap, we present the first data and computation efficient backdoor attacks towards multimodal learning. Our solution consists of two innovations. First, we propose a novel backdoor gradient-based score (BAGS), which can accurately quantify the contribution of each data sample to the backdoor learning at a very early training stage. Therefore, it can greatly save time and computational resources for the attacker. Second, we introduce a searching strategy with two attack modes to efficiently determine the optimal poisoning modalities and data samples.Our methodology leads to the following research outcomes. First, we comprehensively evaluate the proposed solution over state-of-the-art multimodal tasks, models, datasets and settings, to verify its effectiveness, efficiency and transferability. For instance, we only need to poison 0.005% of training samples to attack the Visual Question Answering task with the success rate of >96%. For the Audio Video Speech Recognition task, we poison 0.05% of samples to achieve the success rate of >93%. Second, we disclose several interesting findings during our experiments: (1) poisoning all modalities is not always better than individual ones, sometimes even making the attack worse; (2) modality competition and complementarity coexist in multimodal learning backdoor attacks; (3) A dominant modality in multimodal learning may not dominate the backdoor attacks. We hope this work will spur future research in improving the security of multimodal learning. Code is available at https://github.com/multimodalbags/BAGS_Multimodal.

Federated learning (FL) has been widely applied in Internet of Things (IoT). However, two security problems hinder the proliferation of FL in practical IoT, i.e., privacy leakage and poisoning attacks. To address these problems, various approaches have been proposed from different perspectives. Nevertheless, there remain two critical challenges: 1) how to establish a unified framework for protecting privacy and defending against poisoning attacks and 2) how to implement such methods in the flexible computing architecture of fog computing. In this article, we propose CROSSBEAM, a comprehensive scheme that provides both defense against poisoning attacks and privacy protection for FL in fog computing. Specifically, we construct frameworks to defend against poisoning attacks under both independent and identically distributed (IID) and non-IID settings. Meanwhile, we establish an actively secure framework to protect users’ privacy, building a bridge between privacy protection and poisoning defense. Our CROSSBEAM allows multiple fog nodes and users to collaboratively achieve the FL training. Besides, it can effectively alleviate the negative impact caused by poisoning attacks, meanwhile, users’ data confidentiality can still be guaranteed, even if multiple active fog nodes collude with each other to infer users’ privacy. Additionally, our scheme is of robustness to participants (fog nodes and users) being off-line during the training process. Moreover, benefited from the superiorities of our hierarchical mechanism and secure framework, our scheme can perform with high efficiency. We present rigorous security proof and extensive performance analysis for our CROSSBEAM.

As a promising service, Machine Learning as a Service (MLaaS) provides personalized inference functions for clients through paid APIs. Nevertheless, it is vulnerable to model extraction attacks, in which an attacker can extract a functionally-equivalent model by repeatedly querying the APIs with crafted samples. While numerous works have been proposed to defend against model extraction attacks, existing efforts are accompanied by limitations and low comprehensiveness. In this article, we propose AMAO, a comprehensive defense framework against model extraction attacks. Specifically, AMAO consists of four interlinked successive phases: adversarial training is first exploited to weaken the effectiveness of model extraction attacks. Then, malicious query detection is used to detect malicious queries and mark malicious users. After that, we develop a label-flipping poisoning attack to instruct the adaptive query responses to malicious users. Besides, the image pHash algorithm is employed to ensure the indistinguishability of the query responses. Finally, the perturbed results are served as a backdoor to verify the ownership of any suspicious model. Extensive experiments demonstrate that AMAO outperforms existing defenses in defending against model extraction attacks and is also robust against the adaptive adversary who is aware of the defense.

Class incremental learning from a pre-trained DNN model is gaining lots of popularity. Unfortunately, the pre-trained model also introduces a new attack vector, which enables an adversary to inject a backdoor into it and further compromise the downstream models learned from it. Prior works proposed backdoor attacks against the pre-trained models in the transfer learning scenario. However, they become less effective when the adversary does not have the knowledge of the downstream tasks or new data, which is more practical and considered in this paper. To this end, we design the first latent backdoor attacks against incremental learning. We propose two novel techniques, which can effectively and stealthily embed a backdoor into the pre-trained model. Such backdoor can only be activated when the pre-trained model is extended to a downstream model with incremental learning. It has a very high attack success rate, and is able to bypass existing backdoor detection approaches. Extensive experiments confirm the effectiveness of our attacks over different datasets and incremental learning methods, as well as strong robustness against state-of-the-art backdoor defense mechanisms includingNeural Cleanse,Fine-PruningandSTRIP.

Decentralized multi-client functional encryption for inner product (DMCFE-IP) enables efficient joint functional computation of private inputs in a secure manner without a trusted third party, which has found successful applications, including distributed statistical analysis and machine learning. However, existing DMCFE-IP schemes suffer several drawbacks, such as lack of support for client dropout, requiring cross-client communication for key generation, and poor efficiency and scalability. To address these issues, we propose an efficient and scalable DMCFE-IP, which supports client dropout and non-interactive decentralized partial decryption key generation. Our scheme mainly exploits appropriate underlying cryptographic primitives, including multi-client functional encryption, digital signature, key agreement, secret sharing, and symmetric encryption, with careful integration to achieve the aforementioned two functionalities. We then extend this scheme to enable privacy-preserving federated learning (PPFL) for the cross-silo scenrio. We provide formal security proof for our scheme and evaluate our DMCFE-IP-based PPFL on several real-world datasets. Compared with the state-of-the-art methods, our approach achieves a speedup of 6.12$\sim 43.36\times$in running time.

In this paper, we presentVerifyML, the first secure inference framework to check the fairness degree of a given Machine learning (ML) model.VerifyMLis generic and is immune to any obstruction by the malicious model holder during the verification process. We rely on secure two-party computation (2 PC) technology to implementVerifyML, and carefully customize a series of optimization methods to boost its performance for both linear and nonlinear layer execution. Specifically, (1)VerifyMLallows the vast majority of overhead to be performed offline, thus meeting the low latency requirements for online inference. (2) To speed up offline preparation, we first design novel homomorphic parallel computing techniques to accelerate the authenticated Beaver's triple (including matrix- vector and convolution triples) generation procedure. It achieves up to$1.7\times$computation speedup and gains at least$10.7\times$less communication overhead compared to state-of-the-art work. (3) We also present a new cryptographic protocol to evaluate the activation functions of non-linear layers, which is$4\times$–$42\times$faster and has$\gt 48\times$less communication than the existing 2 PC protocol against malicious parties. In fact,VerifyMLeven beats the state-of-the-art semi-honest ML secure inference system! We provide a formal theoretical analysis forVerifyMLsecurity and demonstrate its performance superiority on mainstream ML models including ResNet-18 and LeNet.

In this paper, we study the problem of secure ML inference against a malicious client and a semi-trusted server such that the client only learns the inference output while the server learns nothing. This problem is first formulated by Lehmkuhlet al.with a solution (MUSE, Usenix Security'21), whose performance is then substantially improved by Chandranet al.'s work (SIMC, USENIX Security'22). However, there still exists a nontrivial gap in these efforts towards practicality, giving the challenges of overhead reduction and secure inference acceleration in an all-round way. Based on this, we propose SIMC 2.0, which complies with the underlying structure of SIMC, but significantly optimizes both the linear and non-linear layers of the model. Specifically, (1) we design a new coding method for parallel homomorphic computation between matrices and vectors. (2) We reduce the size of the garbled circuit (GC) (used to calculate non-linear activation functions,e.g., ReLU) in SIMC by about two thirds. Compared with SIMC, our experiments show that SIMC 2.0 achieves a significant speedup by up to$17.4\times$for linear layer computation, and at least$1.3\times$reduction of both the computation and communication overhead in the implementation of non-linear layers under different data dimensions. Meanwhile, SIMC 2.0 demonstrates an encouraging runtime boost by$2.3\sim 4.3\times$over SIMC on different state-of-the-art ML models.

This work studies secure inference on Binary Neural Networks (BNNs), which have binary weights and activations as a desirable feature. Although previous works have developed secure methodologies for BNNs, they still have performance limitations and significant gaps in efficiency when applied in practice. We present SecBNN, an efficient secure two-party inference framework on BNNs. SecBNN exploits appropriate underlying primitives and contributes efficient protocols for the non-linear and linear layers of BNNs. Specifically, for non-linear layers, we introduce a secure sign protocol with an innovative adder logic and customized evaluation algorithms. For linear layers, we propose a new binary matrix multiplication protocol, where a divide-and-conquer strategy is provided to recursively break down the matrix multiplication problem into multiple sub-problems. Building on top of these efficient ingredients, we implement and evaluate SecBNN over two real-world datasets and various model architectures under LAN and WAN. Experimental results show that SecBNN substantially improves the communication and computation performance of existing secure BNN inference works by up to$29 \times $and$14 \times $, respectively.

Transfer learning, successful in knowledge translation across related tasks, faces a substantial privacy threat from membership inference attacks (MIAs). These attacks, despite posing significant risk to ML model’s training data, remain limited-explored in transfer learning. The interaction between teacher and student models in transfer learning has not been thoroughly explored in MIAs, potentially resulting in an under-examined aspect of privacy vulnerabilities within transfer learning. In this paper, we propose a new MIA vector against transfer learning, to determine whether a specific data point was used to train the teacher model while only accessing the student model in a white-box setting. Our method delves into the intricate relationship between teacher and student models, analyzing the discrepancies in hidden layer representations between the student model and its shadow counterpart. These identified differences are then adeptly utilized to refine the shadow model’s training process and to inform membership inference decisions effectively. Our method, evaluated across four datasets in diverse transfer learning tasks, reveals that even when an attacker only has access to the student model, the teacher model’s training data remains susceptible to MIAs. We believe our work unveils the unexplored risk of membership inference in transfer learning.

There is an increasing tendency to fine-tune large-scale pre-trained language models (LMs) using small private datasets to improve their capability for downstream applications. In this paper, we systematically analyze the pre-train and then fine-tune the process of generative LMs and show that the fine-tuned LMs would leak sensitive keywords of the private datasets even without any prior knowledge of the downstream tasks. Specifically, we propose a novel and efficient keyword inference attack framework to accurately and maximally recover sensitive keywords. Owing to the fine-tuning process, pre-trained and fine-tuned models might respond differently to identical input prefixes. To identify potential sensitive sentences for training the fine-tuend LM, we introduce a contrast difference score that assesses the response variations between a pre-trained LM and its corresponding fine-tuned LM. Following this, we iteratively fine-tune the pre-trained model using these sensitive sentences to minimize the disparity between the target model and the pre-trained model, thereby maximizing the number of inferred sensitive keywords. We implement two types of keyword inference attacks (i.e., domain and private) according to our framework and conduct comprehensive experiments on three downstream applications to evaluate the performance. The experimental results demonstrate that our domain keyword inference attack achieves a precision of 85%, while our private keyword inference attack can extract highly sensitive personal information for a significant number of individuals (approximately 0.3% of all customers in the private fine-tuning dataset, which contains 40,000 pieces of personal information).

Federated Learning (FL) suffers from low convergence and significant accuracy loss due to local biases caused by non-Independent and Identically Distributed (non-IID) data. To enhance the non-IID FL performance, a straightforward idea is to leverage the Generative Adversarial Network (GAN) to mitigate local biases using synthesized samples. Unfortunately, existing GAN-based solutions have inherent limitations, which do not support non-IID data and even compromise user privacy. To tackle the above issues, we propose a GAN-based unbiased FL scheme, calledFlGan, to mitigate local biases using synthesized samples generated by GAN while preserving user-level privacy in the FL setting. Specifically,FlGanfirst presents a federated GAN algorithm using the divide-and-conquer strategy that eliminates the problem of model collapse in non-IID settings. To guarantee user-level privacy,FlGanthen exploits Fully Homomorphic Encryption (FHE) to design the privacy-preserving GAN augmentation method for the unbiased FL. Extensive experiments show thatFlGanachieves unbiased FL with$10\%-60\%$accuracy improvement compared with two state-of-the-art FL baselines (i.e., FedAvg and FedSGD) trained under different non-IID settings. The FHE-based privacy guarantees only cost about 0.53% of the total overhead inFlGan.

As smartphones proliferate, secure and user-friendly authentication methods are increasingly critical. Existing behavioral biometrics, however, are often compromised by behavior variability, leading to poor authentication accuracy and an unsatisfactory user experience. To fill this gap, we proposeBioHold, a new robust and reliable user authentication method, fusing finger behavior and hand geometry, captured via a smartphone's multitouch screen during natural holding gestures. It synergistically fuses behavioral and physiological biometrics. In contrast to traditional methods that require restrictive, unnatural user patterns, our approach utilizes a stable, natural gesture for authentication, effectively mitigating behavior variability. It enables one-handed authentication through familiar smartphone-holding and unlocking gestures. During this interaction, hand geometry and behavioral characteristics are recorded for subsequent authentication. We evaluate our method using a dataset collected from 20 subjects, demonstrating its resilience against behavioral variability over time while maintaining a high level of distinctiveness. With only 10 training samples, our method achieves an equal error rate of 3.59%, which improves to 1.25% with 40 training samples. Importantly, our method is resistant to common security threats such as zero-effort attacks, smudge attacks, and shoulder surfing attacks. A usability study confirms the method's high user acceptance, as measured by the system usability score.

Cloud computing is the widespread acceptance of a promising paradigm offering a substantial amount of storage and data services on demand. To preserve data confidentiality, many cryptosystems have been introduced. However, current solutions are incompatible with the resource-constrained end-devices because of a variety of vulnerabilities in terms of practicality and security. In this paper, we propose a practical and secure data-sharing system by introducing a new design of attribute-based encryption with verifiable outsourced decryption (VO-ABE for short). Our system offers: (1) data sharing at a fine-grained level; (2) a scalable key issuing protocol without any secure channel; (3) a verifiable outsourced decryption mechanism for resource-constrained end-devices against the malicious cloud service provider; and (4) adaptive security against the real-world attacks. To formalize our solution with cryptographic analysis, we present the formal definition of VO-ABE and its concrete construction with provable security. In particular, our design leverages the techniques of the traditional ABE, verifiable outsourced decryption, and randomness extractor to support fine-grained access control, cost-effective data sharing, and security assurance with high entropy. Moreover, our design is provably secure in the adaptive model under the standard assumption, which offers a stronger security guarantee since the state-of-the-art solution is selectively secure under the non-standard assumption and suffers from a variety of real-world attacks. The implementation and evaluation demonstrate that our solution enjoys superior functionality and better performance than the relevant solutions. More importantly, our solution is compatible with the resource-constrained end-devices since the decryption mechanism takes around 1.1ms and is 22.7x faster than the state-of-the-art solution.

Recently, outsourcing machine learning inference services to the cloud has become increasingly popular. The inference process, however, remains an open question onhow to effectively protect the model owner's proprietary model, the user's sensitive data, and prediction results. In this work, we propose an efficient and comprehensive privacy-preserving framework for outsourcing Gradient Boosting Decision Tree (GBDT) inference utilizing pseudorandom function and additively homomorphic encryption. Specifically, we first design a transformation method for GBDT to protect the node and structure privacy of the owner's model. On top of the protected model, we further propose customized comparison and random trees permutation protocols, which substantially boost the computation and reduce the communication cost of the outsourcing inference, while preventing the user from inferring privacy associated with GBDT. Besides, we provide rigorous security analysis, and extensive experiments on 7 real-world datasets and various models demonstrating that our scheme achieves up to 36 times less runtime and 69 times less communication compared to the state-of-the-arts.

Backdoor attacks against neural networks have been intensively investigated, where the adversary compromises the integrity of the victim model, causing it to make wrong predictions for inference samples containing a specific trigger. To make the trigger more imperceptible and human-unnoticeable, a variety of stealthy backdoor attacks have been proposed, some works employ imperceptible perturbations as the backdoor triggers, which restrict the pixel differences of the triggered image and clean image. Some works use special image styles (e.g., reflection, Instagram filter) as the backdoor triggers. However, these attacks sacrifice the robustness, and can be easily defeated by common preprocessing-based defenses. This paper presents a novel color backdoor attack, which can exhibit robustness and stealthiness at the same time. The key insight of our attack is to apply a uniform color space shift for all pixels as the trigger. This global feature is robust to image transformation operations and the triggered samples maintain natural-looking. To find the optimal trigger, we first define naturalness restrictions through the metrics of PSNR, SSIM and LPIPS. Then we employ the Particle Swarm Optimization (PSO) algorithm to searchfor the optimal trigger that can achieve high attack effectiveness and robustness while satisfying the restrictions. Extensive experiments demonstrate the superiority of PSO and the robustness of color backdoor against different main-stream backdoor defenses.

Heterogeneous federated learning (HFL) enables clients with different computation and communication capabilities to collaboratively train their own customized models via a query-response paradigm on auxiliary datasets. However, such a paradigm raises serious privacy concerns due to the leakage of highly sensitive query samples and response predictions. We put forth GuardHFL, the first-of-its-kind efficient and privacy-preserving HFL framework. GuardHFL is equipped with a novel HFL-friendly secure querying scheme built on lightweight secret sharing and symmetric-key techniques. The core of GuardHFL is two customized multiplication and comparison protocols, which substantially boost the execution efficiency. Extensive evaluations demonstrate that GuardHFL significantly outperforms the alternative instantiations based on existing state-of-the-art techniques in both runtime and communication cost.

Federated learning (FL) is a distributed machine learning technology that preserves data privacy. However, it has been shown to be vulnerable to gradient leakage attacks (GLA), which can reconstruct private training data from public gradients with an overwhelming probability. Nevertheless, these attacks either require modification of the FL model (analytics-based) or take a long time to converge (optimization-based) and fail in dealing with highly compressed gradients in practical FL systems. In this paper, we pioneer a generation-based GLA method called FGLA that can reconstruct batches of user data, forgoing the optimization process. Specifically, we design a feature separation technique that extracts the feature of each data in a batch and then generates user data directly. Extensive experiments on multiple image datasets demonstrate that FGLA can reconstruct user images in milliseconds with a batch size of 256 from highly compressed gradients (0.8% compression ratio or higher), thus substantially outperforming state-of-the-art methods.

Deep learning has shown impressive performance in numerous applications. However, recent studies have found that deep learning models are vulnerable to adversarial attacks, where the attacker adds imperceptible perturbations into benign samples to induce misclassifications. Adversarial attacks in the digital domain focus on constructing imperceptible perturbations. However, they are always less effective in the physical world because the perturbations may be destroyed when captured by the camera. Most physical adversarial attacks require adding invisible adversarial features (e.g., a sticker or a laser) to the target object, which may be noticed by human eyes. In this work, we propose to employ image transformation to generate more natural adversarial samples in the physical world. Concretely, we propose two attack algorithms to satisfy different attack goals:Efficient-AATRemploys a greedy strategy to generate adversarial samples with fewer queries;Effective-AATRemploys an adaptive particle swarm optimization algorithm to search for the most effective adversarial samples within the given the number of queries. Extensive experiments demonstrate the superiority of our attacks compared with state-of-the-art adversarial attacks under mainstream defenses.

The population of cloud computing greatly facilitates the sharing of explosively generated image today. While benefiting from the convenient of cloud, the privacy protection mechanism that commonly applied in cloud service makes the spreading of illegal and harmful data very hard to be detected or controlled. Such a realistic threat should be seriously treated, yet is largely overlooked in the literature. To address this issue, we propose the first cloud service framework that can simultaneously provide privacy protection and content regulation for the cloud storage image. In specific, we design a secure multi-party computation (MPC) protocol to protect the data privacy via random projection. By leveraging the distance preserving properties residing in random projection, we propose a privacy-preserving principal component analysis (PCA)-based recognition approach over the random projection domain to achieve content matching while respecting the data privacy. To facilitate the efficiency, we implement our system under the compressive sensing (CS) framework. Due to the compression effect of CS, the proposed cloud service can achieve remarkable reduction on the computation and communication complexity of the content matching process. Theoretical analysis and experimental results both show that our system can achieve privacy assurance and acceptable recognition performance, while with high efficiency.

Benefiting from the powerful computing and storage capabilities of cloud services, data sharing in the cloud has been permeated across various applications including social networks, e-health and crowdsourcing transportation system. Intuitively, outsourcing data to untrusted cloud commonly raises concerns about data privacy breaches. To combat this, one approach is exploiting Broadcast Based Searchable Encryption (BBSE) for secure data sharing. Nevertheless, the latest proposed BBSE is still defective in either security or efficiency. In this article, we propose ESPD, an Efficient, Scalable and Privacy-preserving Data sharing framework over encrypted cloud dataset. Different from previous works, ESPD supports sharing target data to multiple users with distinct secret keys, and keeps a constant ciphertext length with the changes of the amount of system users. This feature significantly improves search efficiency and makes ESPD scalable in real-world scenarios. We show a formal analysis to prove the security of ESPD in terms of file privacy, keyword privacy and trapdoor privacy. Also, extensive experiments on real-world dataset are conducted to indicate the desirable performance of ESPD compared to other similar schemes.

Decentralized image classification plays a key role in various scenarios due to its attractive properties, including tolerating high network latency and less prone to single-point failures. Unfortunately, training such a decentralized image classification model is more vulnerable to data privacy leaks compared to other distributed training frameworks. Existing efforts exclusively use differential privacy as the cornerstone to alleviate the threat to data privacy. However, differential privacy is implemented at the expense of accuracy, which goes against our motivation for designing an image classification model without loss of accuracy. To address this problem, we propose D2-MHE, thefirstsecure and efficient decentralized training framework with lossless precision. Inspired by the latest developments in the homomorphic encryption technology, we design a multiparty version of Brakerski-Fan-Vercauteren (BFV), one of the most advanced cryptosystems, and use it to implement private gradient updates of users’ local models. D2-MHE can reduce the communication complexity of general Secure Multiparty Computation (MPC) tasks from quadratic to linear in the number of users, making it very suitable and scalable for large-scale decentralized learning systems. Moreover, D2-MHE provides strict semantic security protection even if the majority of users are dishonest with collusion. We conduct extensive experiments on MNIST, CIFAR-10, and ImageNet to demonstrate the superiority of D2-MHE. Experimental results show that D2-MHE achieves up to$5.5\times $reduction in computation overhead, and at least$12\times $reduction in communication overhead compared to existing schemes.

In this paper, we address the problem of privacy-preserving federated neural network training with$N$users. We presentHercules, an efficient and high-precision training framework that can tolerate collusion of up to$N-1$users.Herculesfollows the POSEIDON framework proposed by Sav et al. (NDSS’21), but makes a qualitative leap in performance with the following contributions: (i) we design a novel parallel homomorphic computation method for matrix operations, which enables fast Single Instruction and Multiple Data (SIMD) operations over ciphertexts. For the multiplication of two$h\times h$dimensional matrices, our method reduces the computation complexity from$O(h^{3})$to$O(h)$. This greatly improves the training efficiency of the neural network since the ciphertext computation is dominated by the convolution operations; (ii) we present an efficient approximation on the sign function based on the composite polynomial approximation. It is used to approximate non-polynomial functions (i.e.,ReLUandmax), with the optimal asymptotic complexity. Extensive experiments on various benchmark datasets (BCW, ESR, CREDIT, MNIST, SVHN, CIFAR-10 and CIFAR-100) show that compared with POSEIDON,Herculesobtains up to 4% increase in model accuracy, and up to$60\times$reduction in the computation and communication cost.

To reduce the cost of human and material resources and improve the collaborations among medical systems, research laboratories and insurance companies for healthcare researches and commercial activities, electronic medical records (EMRs) have been proposed to shift from paperwork to friendly shareable electronic records. To take advantage of EMRs efficiently and reduce the cost of local storage, EMRs are usually outsourced to the remote cloud for sharing medical data with authorized users. However, cloud service providers are untrustworthy. In this paper, we propose an efficient, secure, and flexible EMR sharing system by introducing a novel cryptosystem called dual-policy revocable attribute-based encryption and tamper resistance blockchain technology. Our proposed system enables EMRs to be shared at a fine-grained level and allows data users to detect any unauthorized manipulation. Moreover, the key generation center can revoke malicious users without affecting the honest users. We provide the formal security model as well as the concrete scheme with security analysis. The experimental simulation and experimental analysis of our proposed scheme demonstrate that our proposed system has superior performances to the most relevant solutions.

Privacy-preserving decision trees (DTs) in vertical federated learning are one of the most effective tools to facilitate various privacy-critical applications in reality. However, the main bottleneck of current solutions is their huge overhead, mainly due to the adoption of communication-heavy bit decomposition to realize complex non-linear operations, such as comparison and division. In this paper, we presentPriVDT, an efficient two-party framework for private vertical DT training and inference in the offline/online paradigm. Specifically, we customize several cryptographic building blocks based on an advanced primitive, Function Secret Sharing (FSS). First, we construct an optimized comparison protocol to improve the efficiency via reducing the invocation of FSS evaluations. Second, we devise an efficient and privacy-enhanced division protocol without revealing the range of divisors, which utilizes the above comparison protocol and more importantly new designed FSS-based secure range and digital decomposition protocols. Besides, we further reduce the overhead of linear operations by employing lightweight pseudorandom function-based Beaver’s triple techniques. Building on the above efficient components, we implement thePriVDTframework and evaluate it on 5 real-world datasets on both LAN and WAN. Experimental results show that the end-to-end runtime ofPriVDToutperforms the prior art by$42 \sim 510\times $on LAN and$16 \sim 70\times $on WAN. Moreover,PriVDTprovides comparable accuracy to the non-private setting.

The cloud-based data sharing technology with cryptographic primitives enables data owners to outsource data into paradigms and privately share information with arbitrary recipients without geographic barriers. However, we argue that most of existing efforts for outsourced data sharing are either inefficient, inflexible, or incompletely secure due to the following problems: (1) lack of efficient strategies for dynamically designating target ciphertexts to multiple recipients; (2) how to hide the identity of the recipient and (3) how to verify the correctness of outsourced ciphertext transformation without any denial. To the best of our knowledge, no previous work has thoroughly explored the above three issues, motivating us to design such an efficient and comprehensively secure outsourced data sharing mechanism. We design VF-PPBA, the first Verifiable, Fair and Privacy-preserving Broadcast Authorization framework for flexible data sharing in clouds. In more detail, we first invent a new primitive, privacy-preserving multi-recipient broadcast proxy re-encryption (PPMR-BPRE), which enables the authorization of a given ciphertext to different recipients with efficient ciphertext transformation, and further guarantees that any malicious adversary deduces nothing about the identity of the recipient. Then, we present VF-PPBA for flexible data sharing with PPMR-BPRE as the underlying structure, which in addition to inheriting all the functionalities of PPMR-BPRE, is capable of supporting the verifiability of the outcome correctness of the outsourced conversion task, and being immune to the malicious accusation if the outsourcing outcome is correctly completed. We formalize the adversarial models and render comprehensively strict security proofs to prove the security of our proposed solutions. Its performance is also validated via experimental simulations to showcase the practicability and effectiveness.

Data sharing technologies enable users to outsource data and privately share information with arbitrary recipients without geographic barriers. However, existing efforts for secure data sharing are either inflexible, insufficiently-secure or inefficient. In this paper, we invent PS-ME, the first Privacy-aware and Security-enhanced efficient Matchmaking Encryption (ME) for flexible data sharing. To be more specific, we first formulate an identity-based broadcast matchmaking encryption (IB-BME) for one-to-many data sharing, which enables both participants to specify respective access policies to the encrypted data, such that the data can be revealed by multiple recipients in the case that both access policies are satisfied. In IB-BME, a general matchmaking transformation solution realizing one-to-many sharing is initialized. We also formulate the PS-ME with the general matchmaking transformation solution of IB-BME as the underlying approach, which in addition to featuring IB-BME’s all desirable properties, enables efficient decryption, identity anonymity and CCA-security, where we address the open problem of ME regarding CCA-security (raised in CRYPTO’2019). Finally, the comprehensively rigorous security proofs indicate the security of the suggested methodologies. The experimental results are also shown to demonstrate their practicability and effectiveness.

Cloud computing has been widely accepted as a computing paradigm to offer high-quality data services on demand. However, it suffers from various attacks as the cloud service provider and data owners are not in the same trusted domain. To support data confidentiality, existing cloud-based systems apply cryptographic tools to issue the decryption key to data users to share data in a controlled way. However, fine-grained cloud data sharing still faces many challenges, especially when dealing with dynamic user groups. In this paper, we introduce a secure and efficient cloud-based data-sharing system with fine-grained access control and dynamic user groups. Our system enjoys 1) adaptive security in prime-order groups, 2) forward secrecy against revoked user fetches data generated before being revoked, and 3) decryption key exposure resistance against the compromise of the frequently used decryption key, where the previous solutions only concentrate on one or two above-mentioned properties. More specifically, we introduce two timestamp management mechanisms that manage the timestamp in each ciphertext to support dynamic user groups with forward secrecy. By applying the proposed timestamp management mechanisms, we introduce two novel designs of attribute-based encryption schemes with formal definition and security analyses. The proposed schemes are adaptively secure in prime-order groups under a standard assumption and support decryption key exposure resistance. We conduct theoretical analysis and experimental simulation to demonstrate the outperformance of our solutions.

As an increasingly prevalent technology in intelligent autonomous transportation systems, autonomous vehicle platoon has been indicated the ability to significantly reduce fuel consumption as well as heighten highway safety and throughput. However, existing efforts rarely focus on protecting data confidentiality and authenticity in autonomous vehicle platoons. How to ensure secure and high-fidelity platoon-level communication is still in its infancy. This paper makes the first attempt for efficient and secure communication across autonomous vehicle platoons. Specifically, we presentPDSM-FC, the first privacy-preserving data share mechanism with flexible cross-domain authorization over distinctive platoons. The key insight ofPDSM-FCis the design of a new ciphertext conversion technique, which allows a ciphertext to be easily converted into another type of ciphertext, facilitating efficient access by all entities holding the legitimate authorization. As a result,PDSM-FCcan achieve high-fidelity data communication between two unique platoons in ciphertext, so as to complete specific tasks including platoon integration. Rigorous security analysis shows thatPDSM-FCis secure against various attacks such as collusion, forgery and chosen-plaintext attacks. Moreover, theoretical evaluation and extensive experiments demonstrate the practicability ofPDSM-FCin terms of functionality, storage and computation overheads.

With an increasing number of IoT devices being deployed in healthcare, massive amounts of electronic health records (EHRs) are generated and shared in the cloud. To preserve data privacy, one promising data-sharing tool named attribute-based encryption (ABE) has been widely employed. However, it is a challenge to achieve flexible data sharing without loss of confidentiality when authorized users are dynamic. Another challenge is how to guarantee fleet data access time when resource-limited devices are used. In this article, a dynamic access policy ABE (DAP-ABE) system for EHRs in the cloud is proposed. The cloud server can update the access policy without sensitive information, while decryption keys of authorized users do not need to be updated. Authorized users enjoy approximately 0.07 ms data access by outsourcing the majority of the decryption overhead to the cloud server. Furthermore, a verification procedure is embedded in DAP-ABE to check the identities of patients in the data sharing stage, which ensures that no malicious user can upload invalid EHRs. Extensive experiments demonstrate the feasibility and efficiency of the DAP-ABE system.

Robot Operating System (ROS) has been the mainstream platform for research and development of robotic applications. This platform is well-known for lacking security features and efficiency for distributed robotic computations. To address these issues, ROS2 is recently developed by utilizing the Data Distribution Service (DDS) to provide security support. Integrated with DDS, ROS2 is expected to establish the basis for trustworthy robotic ecosystems.

Modern autonomous vehicles adopt state-of-the-art DNN models to interpret the sensor data and perceive the environment. However, DNN models are vulnerable to different types of adversarial attacks, which pose significant risks to the security and safety of the vehicles and passengers. One prominent threat is the backdoor attack, where the adversary can compromise the DNN model by poisoning the training samples. Although lots of effort has been devoted to the investigation of the backdoor attack to conventional computer vision tasks, its practicality and applicability to the autonomous driving scenario is rarely explored, especially in the physical world.

We initiate the study of private inference on Transformer-based models in the client-server setting, where clients have private inputs and servers hold proprietary models. Our main contribution is to provide several new secure protocols for matrix multiplication and complex non-linear functions like Softmax, GELU activations, and LayerNorm, which are critical components of Transformers. Specifically, we first propose a customized homomorphic encryption-based protocol for matrix multiplication that crucially relies on a novel compact packing technique. This design achieves $\sqrt{m} \times$ less communication ($m$ is the number of rows of the output matrix) over the most efficient work. Second, we design efficient protocols for three non-linear functions via integrating advanced underlying protocols and specialized optimizations. Compared to the state-of-the-art protocols, our recipes reduce about half of the communication and computation overhead. Furthermore, all protocols are numerically precise, which preserve the model accuracy of plaintext. These techniques together allow us to implement \Name, an efficient Transformer-based private inference framework. Experiments conducted on several real-world datasets and models demonstrate that \Name achieves $3 \sim 14\times$ less communication and $3 \sim 11\times$ less runtime compared to the prior art.

Federated learning (FL) has emerged as a powerful technology widely applied in Internet of Things (IoT). Recently, researchers have shown an increased interest in privacy-preserving FL withunreliable users. The goal of such works is to achieve private training under ciphertext mode while ensuring that the FL model is mainly derived from the contributions of users with high-quality data. However, the existing work is still in its infancy, and the main challenge faced by many researchers is how to achieve their schemes for meeting the demands of high accuracy and efficiency. To combat that, we propose an efficient privacy-preserving FL (EPPFL) scheme withunreliable users. Specifically, we design a novel scheme to mitigate the negative impact ofunreliable users, where the targeted model is guaranteed to be updated with high-quality data. Through iteratively executing our “Excluding Irrelevant Components” and “Weighted Aggregation,” the FL model converges rapidly while taking limited communication and computation overhead. As a result, not only the model accuracy can be optimized, but also the training efficiency can be improved. Meanwhile, we conduct a secure framework based on the threshold Paillier cryptosystem, which can rigorously protect all user-related private information during the training process. Furthermore, the extensive experiments demonstrate our EPPFL with high-level performance in terms of accuracy and efficiency.

With the increasing traffic volume, enterprises choose to outsource their middlebox services, such as deep packet inspection, to the cloud to acquire rich computational and communication resources. However, since the traffic is redirected to the public cloud, information leakages, such as packet payload and inspection rules, arouse privacy concerns of both middlebox owner and packet senders. To address the concerns, we propose an efficient verifiable deep packet inspection (EV-DPI) scheme with strong privacy guarantees. Specifically, a two-layer architecture is designed and deployed over two non-collusion cloud servers. The first layer fast filters out most of legitimate packets and the second layer supports exact rule matching. During the inspection, the privacy of packet payload and the confidentiality of inspection rules are well preserved. To improve the efficiency, only fast symmetric crypto-systems, such as hash functions, are used. Moreover, the proposed scheme allows the network administrator to verify the execution results, which offers a strong control of outsourced services. To validate the performance of the proposed EV-DPI scheme, we conduct extensive experiments on the Amazon Cloud. Large-scale dataset (millions of packets) is tested to obtain the key performance metrics. The experimental results demonstrate that EV-DPI not only preserves the packet privacy, but also achieves high packet inspection efficiency.

Middlebox is an intermediary network equipment which can be outsourced to remote cloud servers for low-cost and customizable network services, such as load balancer and intrusion detection. A fundamental function of the middlebox is packet inspection, where both the packet header and payload are extracted and analyzed based on inspection rules. However, as the packet may contain sensitive individual or organizational information, it may raise severe privacy concerns without proper countermeasures. In this article, we propose a secure and versatile packet inspection scheme for outsourced middlebox. The proposed scheme builds upon two non-collusion cloud servers, where the first server conducts the inspection task over the ciphertext domain and the second reveal the inspection results. By doing so, the proposed scheme achieves versatile inspection functionalities: range-query-based header inspection and token-based payload inspection, while preserving the privacy of packet header, payload, and inspection rules. Moreover, we identify and address two challenging issues in the state-of-the-art literatures. First, we tailor the design of mis-operation resistant searchable homomorphic encryption (MR-SHE) and somewhat homomorphic encryption in the two-server model, to resistoffline dictionary attack on payload headers. Second, we propose a key management mechanism with compelled access for the middlebox, to achievefine-grained probable cause privacy. We also conduct extensive experiments and compare the results with existing schemes to demonstrate the feasibility of the proposed scheme.

DNA similarity search has been widely applied in human genomic studies including DNA marking, genomic sequencing and genetic disease prediction. Meanwhile, with the explosive growth of data, users are increasingly inclining to store DNA data on the cloud for saving local cost. However, the high sensitivity of DNA data has forced the government to strictly control its acquisition and utilization. One potential solution is to encrypt DNA data before outsourcing them to the cloud. Nevertheless, private DNA similarity query has been an active research issue, state-of-the-art results are still defective in security, functionality, and efficiency. In this article, we propose EFSS, an efficient and fine-grained similarity search scheme over encrypted DNA data. In specific, first, we design an approximation algorithm to efficiently calculate the edit distances between two sequences. Second, we put forward a novel Boolean search strategy to achieve complicated logic queries such as mixed “AND” and “NO” operations on genes. Third, data access control is also supported in our EFSS through a variant of polynomial based design. Moreover, the K-means clustering algorithm is exploited to further improve the efficiency of execution. In the end, security analysis and extensive experiments demonstrate the high performance of EFSS compared with existing schemes.

Image classification is a fundamental artificial intelligence task that labels images into one of some predefined classes. However, training complex image classification models requires a large amount of computation resources and data in order to reach state-of-the-art performance. This demand drives the growth of distributed deep learning, where multiple agents cooperatively train global models with their individual datasets. Among such learning systems, decentralized learning is particularly attractive, as it can improve the efficiency and fault tolerance by eliminating the centralized parameter server, which could be the single point of failure or performance bottleneck. Although the agents do not need to disclose their training image samples, they exchange parameters with each other at each iteration, which can put them at the risk of data privacy leakage. Past works demonstrated the possibility of recovering training images from the exchanged parameters. One common defense direction is to adopt Differential Privacy (DP) to secure the optimization algorithms such as Stochastic Gradient Descent (SGD). Those DP-based methods mainly focus on standalone systems, or centralized distributed learning. How to enforce and optimize DP protection in decentralized learning systems is unknown and challenging, due to their complex communication topologies and distinct learning characteristics. In this paper, we design TOP- DP, a novel solution to optimize the differential privacy protection of decentralized image classification systems. The key insight of our solution is to leverage the unique features of decentralized communication topologies to reduce the noise scale and improve the model usability. (1) We enhance the DP-SGD algorithm with thistopology-awarenoise reduction strategy, and integrate the time-aware noise decay technique. (2) We design two novel learning protocols (synchronous and asynchronous) to protect systems with different network connectivities and topologies. We formally analyze and prove the DP requirement of our proposed solutions. Experimental evaluations demonstrate that our solution achieves a better trade-off between usability and privacy than prior works. To the best of our knowledge, this is the first DP optimization work from the perspective of network topologies.

Federated deep learning has been widely used in various fields. To protect data privacy, many privacy-preservingapproaches have been designed and implemented in various scenarios. However, existing works rarely consider a fundamental issue that the data shared by certain users (calledirregular users) may be of low quality. Obviously, in a federated training process, data shared by manyirregular usersmay impair the training accuracy, or worse, lead to the uselessness of the final model. In this article, we propose PPFDL, a Privacy-Preserving Federated Deep Learning framework withirregular users. In specific, we design a novel solution to reduce the negative impact ofirregular userson the training accuracy, which guarantees that the training results are mainly calculated from the contribution of high-quality data. Meanwhile, we exploit Yao's garbled circuits and additively homomorphic cryptosystems to ensure the confidentiality of all user-related information. Moreover, PPFDL is also robust to users dropping out during the whole implementation. This means that each user can be offline at any subprocess of training, as long as the remaining online users can still complete the training task. Extensive experiments demonstrate the superior performance of PPFDL in terms of training accuracy, computation, and communication overheads.

Healthcare Internet-of-Things (IoT) enables lightweight devices to observe patients’ vital signals and outsource them to a remote cloud to enjoy flexible data sharing. However, it faces many security threats as the outsourced data is no longer physically controlled by data owners, and the cloud that hosts the outsourced data is not fully trusted. Many privacy protection technologies have been adopted to solve this problem, among which cryptographic mechanisms have become one of the most promising tools. Unfortunately, current cryptographic mechanisms in healthcare IoT mainly suffer from the following challenges: 1) dynamic user groups for managing users’ accessibility; 2) efficient revocation mechanism to mitigate the burden during user revocation; 3) forward and backward secrecy to ensure session independence in the presence of session key leakage; 4) revocable storage to prevent data users from learning any unauthorized data even the data is authorized before; and 5) information manipulation during data transmission. In this article, we introduce a practical and secure system to address the above problems. Our system provides fine-grained access control with dynamic user groups for optimizing scalability and functionality. We prove that our system is secure against numerous real-world threats. Extensive comparison and experimental analysis demonstrate that our system enjoys superior performance than the state-of-the-art solutions.

Cloud-fog computing is a novel paradigm to extend the functionality of cloud computing to provide a variety of on-demand data services via the edge network. Many cryptographic tools have been introduced to preserve data confidentiality against the untrustworthy network and cloud servers. However, how to efficiently identify and retrieve useful data from a large number of ciphertexts without a costly decryption mechanism remains a challenging problem. In this article, we introduce a cloud-fog-device data sharing system (CFDS) with data confidentiality and data source identification simultaneously based on a new cryptographic primitive named matchmaking attribute-based encryption (MABE) by extending matchmaking encryption in CRYPTO’19. Our solution offers a secure fine-grained bilateral access control that includes (1) fine-grained sender access control, (2) fine-grained receiver access control, (3) sender privacy, and (4) performance optimization via outsourcing data source identification to fog nodes. We give the formal definition and security models of MABE, and present a concrete construction with formal security proofs. We also offer a detailed security analysis of our proposed CFDS against real-world security threats. The extensive comparison and experimental simulation demonstrate that, by immigrating heavy workload to fog nodes, our scheme has better functionalities and performances than the most related solutions.

Aiming for minimizing passengers waiting time and vehicles vacancy rate, online car-hailing service systems with fog computing has been deployed in various scenarios. In this paper, we focus on addressing the security and privacy issues in such a promising system by customizing a new cryptographic primitive to provide the following security guarantees: (1) private, fine-grained and bilateral order matching between passengers and drivers; (2) authenticity verification of passengers orders in the form of ciphertext, and (3) temporal assurance of passengers’ ciphertext orders. To the best of our knowledge, no previous system has been designed to meet all three requirements. Existing cryptographic primitives (including forward/puncturable encryption (FE/PE) and attribute based matchmaking encryption (AB-ME)) may be leveraged to partially address some of challenges, but there lacks a comprehensive solution. Moreover, the integration of existing works is hampered by the heterogeneity and the weak coupling between distinct cryptographic primitives. As a result, it is infeasible to directly exploit them for the online car-hailing service. To tackle that, we put forward a new cryptographic primitive called Fine-grained Puncturable Matchmaking Encryption (FP-ME) by modifying AB-ME and incorporating PE technology. FP-ME can simultaneously implement fine-grained and bilateral order matching, the authenticity of passengers orders, and meeting the time constraint of passengers orders. We formalize the adversarial models for the proposed FP-ME and then present rigorous security analysis to prove the security of the proposed system. Additionally, we study performance of the system via simulations to demonstrate its practicability and effectiveness in the real-world applications.

The effectiveness of state-of-the-art deep learning (DL) models has empowered the development of industrial Internet of things (IIoT). Recently, considering resource-constrained and privacy-required IIoT devices, collaborative inference has been proposed, which splits DL models and deploys them in IIoT devices and an edge server separately. However, in this article, we argue that there are still severe privacy vulnerabilities in collaborative inference systems. And we devise the first membership inference attack (MIA) against collaborative inference, to infer whether a particular data sample is used for training the model of IIoT systems. Existing MIAs either assume full access to the systems’ APIs or availability of the target model's parameters, which is not applicable in realistic IIoT environments. In contrast to prior works, we proposetransfer-inheritshadow learning and thus relax these key assumptions. We evaluate our attack on different datasets and various settings, and the results show it has high effectiveness.

Federated learning (FL) has demonstrated tremendous success in various mission-critical large-scale scenarios. However, such promising distributed learning paradigm is still vulnerable to privacy inference and byzantine attacks. The former aims to infer the privacy of target participants involved in training, while the latter focuses on destroying the integrity of the constructed model. To mitigate the above two issues, a few works recently explored unified solutions by utilizing generic secure computation techniques and common byzantine-robust aggregation rules, but there are two major limitations: 1) they suffer from impracticality due to efficiency bottlenecks, and 2) they are still vulnerable to various types of attacks because of model incomprehensiveness.

The rapid development of machine learning technology has prompted the applications of Automatic Speech Recognition(ASR). However, studies have shown that the state-of-the-art ASR technologies are still vulnerable to various attacks, which undermines the stability of ASR destructively. In general, most of the existing attack techniques for the ASR model are based on white box scenarios, where the adversary uses adversarial samples to generate a substituted model corresponding to the target model. On the contrary, there are fewer attack schemes in the black-box scenario. Moreover, no scheme considers the problem of how to construct the architecture of the substituted models. In this paper, we point out that constructing a good substituted model architecture is crucial to the effectiveness of the attack, as it helps to generate a more sophisticated set of adversarial examples. We evaluate the performance of different substituted models by comprehensive experiments, and find that ensemble substituted models can achieve the optimal attack effect. The experiment shows that our approach performs attack over 80% success rate (2% improvement compared to the latest work) meanwhile maintaining the authenticity of the original sample well.

Proof of replication technique has been widely used to verify whether the cloud service providers (CSPs) store multiple replications of a file with dedicated and unique storage space, which effectively prevents CSPs from colluding and storing only one copy of the file. In this field, many representative schemes have been proposed and applied to various scenarios. However, most of the existing schemes are based on the timing assumption (i.e., the verifier rejects the proof of replication if the prover's response is timeout) and do not explicitly consider the problem of batch verification and fault localization. This will bring unnecessary computational overhead to the verifier and reduce the efficiency of batch auditing. To address the above problems, we propose a verifiable proof of replication scheme with fast fault localization and high efficiency. By integrating incompressible encoding and homomorphic linear authenticator, our scheme can effectively audit the integrity of file replications without timing assumptions. To support batch verification and fault localization, we propose a reversed signature aggregation tree (Rev-tree) by integrating the quick binary search and exponent testing. Compared with the traditional binary tree, Rev-tree can further reduce the overhead of batch verification and effectively locate a single fault replication. Moreover, benefit from the property of Rev-tree taking the existing error probability as an estimate of the rest of the tree, our scheme can adjust the verification strategy dynamically to meet with different situations. Finally, security analysis and experimental results show that our scheme is secure and efficient in proof of replication and fast fault localization.

The emerging of healthcare Industrial Internet of Things (HealthIIoT) cannot only facilitate high-quality care services for patients but also enable efficient telemedicine platform for healthcare practitioners. However, it faces several fundamental security and privacy challenges, such as secure fine-grained data delivery, privacy preserving keyword-based ciphertext retrieval, malicious key delegation, and efficiency of the system. To combat these issues, we propose a privacy-aware and traceable fine-grained system (PTFS) for secure data delivery in cloud-assisted HealthIIoT. Compared to the existing solutions that only implement some of the preceding features, the proposed solution enables secure fine-grained data delivery, privacy-preserving data retrieval, efficient encryption and decryption operations, and trace of malicious key delegation simultaneously. For security analysis, rigorous proofs of the proposed scheme are provided to prove its security. In addition, extensive simulations and experiments are conducted for performance evaluation, which demonstrate the feasibility and effectiveness of PTFS.

Federated learning (FL), as a distributed machine learning setting, has received considerable attention in recent years. To alleviate privacy concerns, FL essentially promises that multiple parties jointly train the model by exchanging gradients rather than raw data. However, intrinsic privacy issue still exists in FL, e.g., user’s training samples could be revealed by solely inferring gradients. Moreover, the emerging poisoning attack also poses a crucial security threat to FL. In particular, due to the distributed nature of FL, malicious users may submit crafted gradients during the training process to undermine the integrity and availability of the model. Furthermore, there exists a contradiction in simultaneously addressing two issues, that is, privacy-preserving FL solutions are dedicated to ensuring gradients indistinguishability, whereas the defenses against poisoning attacks tend to remove outliers based on their similarity. To solve such a dilemma, in this paper, we aim to build a bridge between the two issues. Specifically, we present a privacy-enhanced FL (PEFL) framework that adopts homomorphic encryption as the underlying technology and provides the server with a channel to punish poisoners via the effective gradient data extraction of the logarithmic function. To the best of our knowledge, the PEFL is the first effort to efficiently detect the poisoning behaviors in FL under ciphertext. Detailed theoretical analyses illustrate the security and convergence properties of the scheme. Moreover, the experiments conducted on real-world datasets show that the PEFL can effectively defend against label-flipping and backdoor attacks, two representative poisoning attacks in FL.

Outsourced inference service has enormously promoted the popularity of deep learning, and helped users to customize a range of personalized applications. However, it also entails a variety of security and privacy issues brought by untrusted service providers. Particularly, a malicious adversary may violate user privacy during the inference process, or worse, return incorrect results to the client through compromising the integrity of the outsourced model. To address these problems, we propose SecureDL to protect the model’s integrity and user’s privacy in Deep Neural Networks (DNNs) inference process. In SecureDL, we first transform complicated non-linear activation functions of DNNs to low-degree polynomials. Then, we give a novel method to generate sensitive-samples, which can verify the integrity of a model’s parameters outsourced to the server with high accuracy. Finally, We exploit Leveled Homomorphic Encryption (LHE) to achieve the privacy-preserving inference. We shown that our sensitive-samples are indeed very sensitive to model changes, such that even a small change in parameters can be reflected in the model outputs. Based on the experiments conducted on real data and different types of attacks, we demonstrate the superior performance of SecureDL in terms of detection accuracy, inference accuracy, computation, and communication overheads.

Truth Discovery (TD) is to infer truthful information by estimating the reliability of users in crowdsensing systems. To protect data privacy, many Privacy-Preserving Truth Discovery (PPTD) approaches have been proposed. However, all existing PPTD solutions do not consider a fundamental issue of trust. That is, if the data aggregator (e.g., the cloud server) is not trustworthy, how can an entity be convinced that the data aggregator has correctly performed the PPTD? A "lazy" cloud server may partially follow the deployed protocols to save its computing and communication resources, or worse, maliciously forge the results for some shady deals. In this paper, we propose V-PATD, the first Verifiable and Privacy-Aware Truth Discovery protocol in crowdsensing systems. In V-PATD, a publicly verifiable approach is designed enabling any entity to verify the correctness of aggregated results returned from the server. Since most of the computation burdens are carried by the cloud server, our verification approach is efficient and scalable. Moreover, users' data is perturbed with the principles of local differential privacy. Security analysis shows that the proposed perturbation mechanism guarantees a high aggregation accuracy even if large noises are added. Compared to existing solutions, extensive experiments conducted on real crowdsensing systems demonstrate the superior performance of V-PATD in terms of accuracy, computation and communication overheads.

With the wild applications of machine learning (ML) technology, automatic speech recognition (ASR) has made great progress in recent years. Despite its great potential, there are various evasion attacks of ML-based ASR, which could affect the security of applications built upon ASR. Up to now, most studies focus on white-box attacks in ASR, and there is almost no attention paid to black-box attacks where attackers can only query the target model to get output labels rather than probability vectors in audio domain. In this paper, we propose an evasion attack against ASR in the above-mentioned situation, which is more feasible in realistic scenarios. Specifically, we first train a substitute model by using data augmentation, which ensures that we have enough samples to train with a small number of times to query the target model. Then, based on the substitute model, we apply Differential Evolution (DE) algorithm to craft adversarial examples and implement black-box attack against ASR models from the Speech Commands dataset. Extensive experiments are conducted, and the results illustrate that our approach achieves untargeted attacks with over 70% success rate while still maintaining the authenticity of the original data well.

The widespread use of edge devices in E-Health such as smartphones and wearables means richer electronic health records (EHR) are becoming available. Training deep learning models on these data can effectively improve the quality of healthcare services. Recently, federated learning (FL) has received extensive attention in E-Health because it can train a model by only sharing gradients without disclosing the original EHR of owners. In this case, however, the adversary can still violate EHR owners' privacy based on shared gradients. To mitigate privacy threat, several privacy-preserving FL protocols have been proposed by utilizing different cryptography techniques. Unfortunately, existing privacy-preserving FL schemes do not take into account irrelevant updates, which are useless for the convergence of the global model. This may reduce the predictive accuracy and worse may lead to the uselessness of the final model. In this paper, we propose PFL-IU, an efficient and privacy-preserving FL framework that is compatible with irrelevant updates. Specifically, we first design a communication-efficient secure aggregation protocol by using a non-interactive key generation algorithm. Then we present a sign method to mitigate the negative impact incurred by irrelevant updates, which will accelerate model convergence and improve predictive accuracy. Moreover, PFL-IU is robust to EHR owners' dropout during the whole training phase. Extensive experiments using the real-world dataset demonstrate that PFL-IU can achieve better performance in terms of accuracy, convergence and efficiency.

Electronic health records (EHR), generated in healthcare, contain extensive digital information, such as diagnoses, medications and complications. Recently, many studies have focused on constructing deep learning (DL) models with EHR data to improve the quality of healthcare services. However, in traditional centralized training, the collection of EHR causes serious privacy issues due to vulnerable transmission channels and untrusted DL service providers. An alternative that can mitigate the above privacy threat is federated learning (FL). It enables multiple healthcare institutions to learn a global predictive model by exchanging locally calculated updates without disclosing the private dataset. Unfortunately, the latest studies have shown that the local updates still expose sensitive information about the original training data. While several privacy-preserving FL protocols have been proposed, few prior works focused on energy consumption issues. Specifically, local training requires extensive computational resources, which is prohibitively expensive for resource-limited institutions. To overcome the above problems, we propose PRCL, a Privacy-aware and Resource-saving Collaborative Learning protocol. To reduce the local computational overhead, we design a novel model splitting method that partitions the neural network into three parts and outsources the computationally large middle part to cloud servers. By using the lightweight data perturbation and packed partially homomorphic encryption, PRCL protects the privacy of the original data and labels, as well as the parameters of the model. Moreover, we analyze the security of the proposed protocol, and demonstrate the superior performance of PRCL in terms of accuracy and efficiency.

Cloud-based deep learning (DL) solutions have been widely used in applications ranging from image recognition to speech recognition. Meanwhile, as commercial software and services, such solutions have raised the need for intellectual property rights protection of the underlying DL models. Watermarking is the mainstream of existing solutions to address this concern, by primarily embedding pre-defined secrets in a model's training process. However, existing efforts almost exclusively focus on detecting whether a target model is pirated, without considering traitor tracing. In this paper, we present SecureMark_DL, which enables a model owner to embed a unique fingerprint for every customer within parameters of a DL model, extract and verify the fingerprint from a pirated model, and hence trace the rogue customer who illegally distributed his model for profits. We demonstrate that SecureMark_DL is robust against various attacks including fingerprints collusion and network transformation (e.g., model compression and model fine-tuning). Extensive experiments conducted on MNIST and CIFAR10 datasets, as well as various types of deep neural network show the superiority of SecureMark_DL in terms of training accuracy and robustness against various types of attacks.

Fog-cloud computing promises many new vertical service areas beyond simple data communication, storing, and processing. Among them, distributed deep learning (DDL) across fog-cloud computing environment is one of the most popular applications due to its high efficiency and scalability. Compared with the centralized deep learning, DDL can provide better privacy protection with training only on sharing parameters. Nevertheless, when DDL meets fog-cloud computing, it still faces two major security challenges: 1) how to protect users' privacy from being leaked to other internal participants in the training process and 2) how to guarantee users' identities from being forged by external adversaries. To combat them, several approaches have been proposed via various technologies. Nevertheless, those approaches suffer from drawbacks in terms of security, efficiency, and functionality, and cannot guarantee the legitimacy of participants' identities during training. In this article, we propose a secure and privacy-preserving DDL (SPDDL) for fog-cloud computing. Compared with the state-of-the-art works, our proposal achieves a better tradeoff between security, efficiency, and functionality. In addition, our SPDDL can guarantee the unforgeability of users' identities against external adversaries. Extensive experimental results indicate the practical feasibility and high efficiency of our SPDDL.

As a promising data-driven technology, deep learning has been widely employed in a variety of Internet-of-Things (IoT) applications. Examples include automated navigation, telemedicine, and smart home. To protect the data privacy of deep-learning-based IoT applications, a few privacy-preserving approaches have also been exploited, designed, and implemented in various scenarios. However, state-of-the-art works are still defective in accuracy, efficiency, and functionality. In this article, we propose the privacy-aware and asynchronous deep-learning-assisted IoT applications (PADL), a privacy-aware and asynchronous deep learning framework that enables multiple data collecting sites to collaboratively train deep neural networks (DNNs), while keeping the confidentiality of private data to each other. Specifically, we first design a layerwise importance propagation (LIP) algorithm to quantify the importance of the model's weights held by each site. Then, we present the customized perturbation mechanism, a precise combination of the LIP algorithm and differential privacy mechanism, which helps to make optimal tradeoffs between the availability and privacy of local models. Furthermore, to fully use the computing resources of all sites, for the first time, we propose an advanced asynchronous optimization (AAO) protocol to perform global updates without waiting. Theoretical analysis shows that the PADL is robust to extreme collusion even with only one reliable site while supporting lock-free optimization. Finally, extensive experiments conducted on real-world data sets using TensorFlow library show that the PADL outperforms the existing systems in terms of efficiency and prediction accuracy.

As an emerging training model with neural networks, federated learning has received widespread attention due to its ability to update parameters without collecting users' raw data. However, since adversaries can track and derive participants' privacy from the shared gradients, federated learning is still exposed to various security and privacy threats. In this paper, we consider two major issues in the training process over deep neural networks (DNNs): 1) how to protect user's privacy (i.e., local gradients) in the training process and 2) how to verify the integrity (or correctness) of the aggregated results returned from the server. To solve the above problems, several approaches focusing on secure or privacy-preserving federated learning have been proposed and applied in diverse scenarios. However, it is still an open problem enabling clients to verify whether the cloud server is operating correctly, while guaranteeing user's privacy in the training process. In this paper, we propose VerifyNet, the first privacy-preserving and verifiable federated learning framework. In specific, we first propose a double-masking protocol to guarantee the confidentiality of users' local gradients during the federated learning. Then, the cloud server is required to provide the Proof about the correctness of its aggregated results to each user. We claim that it is impossible that an adversary can deceive users by forging Proof, unless it can solve the NP-hard problem adopted in our model. In addition, VerifyNet is also supportive of users dropping out during the training process. The extensive experiments conducted on real-world data also demonstrate the practical performance of our proposed scheme.

By leveraging deep learning-based technologies, industrial artificial intelligence (IAI) has been applied to solve various industrial challenging problems in Industry 4.0. However, for privacy reasons, traditional centralized training may be unsuitable for sensitive data-driven industrial scenarios, such as healthcare and autopilot. Recently, federated learning has received widespread attention, since it enables participants to collaboratively learn a shared model without revealing their local data. However, studies have shown that, by exploiting the shared parameters adversaries can still compromise industrial applications such as auto-driving navigation systems, medical data in wearable devices, and industrial robots' decision making. In this article, to solve this problem, we propose an efficient and privacy-enhanced federated learning (PEFL) scheme for IAI. Compared with existing solutions, PEFL is noninteractive, and can prevent private data from being leaked even if multiple entities collude with each other. Moreover, extensive experiments with real-world data demonstrate the superiority of PEFL in terms of accuracy and efficiency.

Using searchable encryption (SE), users' data can be outsourced to an untrusted server while ensuring privacy of both the queries and the data. Meanwhile, to efficiently support data updating, dynamic SE (DSE) has also been proposed and applied to a variety of scenarios. However, recent work shows that even with little information leakage on updated keywords, most of existing DSE schemes are also vulnerable to adaptative attacks breaking the privacy of the queries. To address this problem, several privacy-preserving DSE have been exploited to mitigate the two major privacy issues in the data update process: i.e., Forward privacy and Backward privacy. Nevertheless, it is still an open problem to support clients multi-keyword-based searching over dynamic cloud data. In reality, as a promising query requirement, it is assurance that the cost of all participants can be fundamentally reduced by implementing multi-keyword-based querying. To combat that, in this paper, we design the first multi-keyword based search proposals ensuring forward and backward privacy over dynamic cloud data. Specifically, we utilize Symmetric Hidden Vector Encryption (SHVE) as the underlying structure to build multi-keyword search protocol. Then, Bloom filter integrating with pseudo-random function will be further adopted to enhance query efficiency. The security analysis proves the high security of our model, and extensive experiments conducted on real-world data also demonstrate the practical performance of our proposed scheme.

As a promising training model with Neural Network, distributed deep learning has been widely applied in various scenarios, where clients and the cloud server work together only by sharing local gradients and global parameters. However, research has shown that the adversary can still reconstruct the users' private information even if little information is leaked. To address this problem, several approaches of privacy-preserving distributed training have been exploited with existing mature technologies, such as Differential Privacy, Secure Multi-party Computation and Homomorphic Encryption. However, state of-the-art results are still defective in security, functionality and efficiency. In this paper, we propose an Efficient Privacy Preserving Scheme (EPPS) for distributed deep learning. We claim that our solution achieves the best performance tradeoff between security, efficiency and functionality. Specifically, we adopt the threshold Paillier encryption as the underlying structure to construct our secure training model. Hence, the confidentiality of honest users' of local gradients can be guaranteed, even the cloud server colluding with multiple users. In addition, since users are often accidentally offline due to either network environment or equipment damage, our EPPS can also support users exiting at any phases of the entire work process. Further more, we conducted extensive experiments on real-world data to demonstrate the preferable performance of our proposed scheme.

Deep learning has been applied in many areas, such as computer vision, natural language processing and emotion analysis. Differing from the traditional deep learning that collects users' data centrally, federated deep learning requires participants to train the networks on private datasets and share the training results, and hence has more gratifying efficiency and stronger security. However, it still presents some privacy issues since adversaries can deduce users' privacy from local outputs, such as gradients. While the problem of private federated deep learning has been an active research issue, the latest research findings are still inadequate in terms of security, accuracy and efficiency. In this paper, we propose an efficient and privacy-preserving federated deep learning protocol based on stochastic gradient descent method by integrating the additively homomorphic encryption with differential privacy. Specifically, users add noises to each local gradients before encrypting them to obtain the optical performance and security. Moreover, our scheme is secure to honest-but-curious server setting even if the cloud server colludes with multiple users. Besides, our scheme supports federated learning for large-scale users scenarios and extensive experiments demonstrate our scheme has high efficiency and high accuracy compared with non-private model.

As a basic query function, range query has been exploited in many scenarios such as SQL retrieves, location-based services, and computational geometry. Meanwhile, with explosive growth of data volume, users are increasingly inclining to store data on the cloud for saving local storage and computational cost. However, a long-standing problem is that the user's data may be completely revealed to the cloud server because it has full data access right. To cope with this problem, a frequently-used method is to encrypt raw data before outsourcing them, but the availability and operability of data will be reduced significantly. In this paper, we propose an efficient and geometric range query scheme (EGRQ) supporting searching and data access control over encrypted spatial data. We employ secure KNN computation, polynomial fitting technique, and order-preserving encryption to achieve secure, efficient, and accurate geometric range query over cloud data. Then, we propose a novel spatial data access control strategy to refine user's rights in our EGRQ. To improve the efficiency, R-tree is adopted to reduce the searching space and matching times in whole search process. Finally, we theoretically prove the security of our proposed scheme in terms of confidentiality of spatial data, privacy protection of index and trapdoor, and the unlinkability of trapdoors. In addition, extensive experiments demonstrate the high efficiency of our proposed model compared with existing schemes.

We design a Practical and Privacy-Aware Truth Discovery (PPATD) approach in mobile crowd sensing systems, which supports users to go offline at any time while still achieving practical efficiency under working process. More notably, our PPATD is the first solution under single server setting to resolve the problem that users must be online at all times during the truth discovery. Moreover, we design a double-masking with one-time pads protocol to further ensure the strong security of users' privacy even if there is a collusion between the cloud server and multiple users.

Traditional centralized PKIs are vulnerable due to the single point of failure. A feasible solution is to build a decentralized PKI without certificate authority (CA). Web of Trust is the first step toward realizing a decentralized PKI, but it still has some limitations such as missing incentive and leaking user's privacy. Blockchain's numerous desirable properties, such as cryptographical security, decentralized nature and unalterable transaction record, make it a suitable tool to implement a decentralized PKI. However, the latest research findings about blockchain-based PKI are still incompatible with the thin-clients which have limited storage ability to download the entire blockchain. To combat that, we firstly present a Privacy-preserving Thin-client Scheme (PTS) utilizing the idea of k-anonymity, which enables thin-clients to run normally as full node users and protect user's privacy simultaneously. After that, in order to reduce cost, we further propose an Efficient Privacy preserving Thin-client Scheme (EPTS) employing the method of PIR (private information retrieval). Then security analysis and functional comparison are performed to demonstrate the high security and comprehensive functionality of EPTS compared with existing schemes. Finally, extensive experiments are undertaken to confirm that EPTS can reduce computational cost and communication cost impressively.

Range search of spatial data, has been applied in many scenarios such as geometric queries, location-based services, and computational geometry, etc. Due to the increasing amount of spatial data, which are usually outsourced to the cloud for saving storage and computational overhead. However, a common privacy issue is that the cloud server may steal user's sensitive information utilizing its powerful computing advantages. A feasible way of managing this bottleneck is to encrypt spatial data before outsourcing it. Nevertheless, the availability of data will be significantly reduced because of the query difficulty over the encrypted cloud data. In this paper, we propose an Efficient Range Search scheme (EFRS) which can achieve fine- grained query over encrypted spatial data. We original contributions are threefold. First, polynomial fitting technique and orderpreserving encryption are introduced to realize the efficient and fine- grained range query over encrypted cloud data. Then, in order to improve the search efficiency, we exploit the Rtree to significantly decreased the search space. Finally, we theoretically proved the security of our proposed scheme in terms of confidentially of spatial data, privacy protection of index and trapdoor, and the unlinkability of trapdoor. Besides, extensive experiments demonstrate the high efficiency of our proposed model compared with existing schemes.

In big data era, data are usually stored in databases for easy access and utilization, which are now woven into every aspect of our lives. However, traditional relational databases cannot address users' demands for quick data access and calculating, since they cannot process data in a distributed way. To tackle this problem, non-relational databases such as MongoDB have emerged up and been applied in various Scenarios. Nevertheless, it should be noted that most MongoDB products fail to consider user's data privacy. In this paper, we propose a practical encrypted MongoDB (i.e., CryptMDB). Specifically, we utilize an additive homomorphic asymmetric cryptosystem to encrypt user's data and achieve strong privacy protection. Security analysis indicates that the CryptMDB can achieve confidentiality of user's data and prevent adversaries from illegally gaining access to the database. Furthermore, extensive experiments demonstrate that the CryptMDB achieves better efficiency than existing relational database in terms of data access and calculating.

With the rapid development of portable mobile devices, crowd sensing systems have been recognized as a key technology to utilize the data collected by the portable mobile devices towards scalable and flexible mobile services. However, since the information provided by devices may not be reliable, the aggregated results of the collected data may not be accurate. To tackle this challenge, various truth discovery schemes have been proposed. On the other hand, a practical issue of privacy protection is not considered in most existing truth discovery schemes. In this paper, we propose an Efficient Privacy-preserving Truth Discovery (EPTD) in Crowd Sensing Systems, which can protect the privacy of users' observed values and weights in truth discovery process. Finally, we show the performance of our scheme is better than existing models in terms of computation overhead.