Gimbal-stabilized visual tracking is critical for modern autonomous systems such as Unmanned Aerial Vehicles (UAVs). While prior work shows acoustic signals can disturb gimbal internals, the impact of such attacks on real-world applications like UAV tracking and following remains underexplored. Existing demonstrations largely overlook practical challenges for real-world attacks, such as object-motion uncertainty and runtime latency. To bridge this gap, we present Banshee, the first physically realizable attack that induces target switching in UAV visual tracking systems by exploiting acoustic vulnerabilities in gimbal-camera systems. Banshee generates carefully crafted acoustic waveforms that induce optimized adversarial gimbal oscillations, causing directionally biased camera-view drifts that break inter-frame target associations. Consequently, the onboard tracker is driven to switch from the original target to an attacker-selected object with high probability, with occasional target loss. Banshee achieves a 93.6% success rate in simulation across two commercial gimbal systems and five trackers. Real-world benchtop and in-flight black-box attacks against a commercial drone across varied scenarios show an overall 95.5% attack success rate. Our results reveal a practical cross-domain vulnerability between acoustics and vision, highlighting the need for robust designs of gimbal systems and applications. Our code is available at: https://github.com/U1ltra/Banshee.

An important paradigm in 3D object detection is the use of multiple modalities to enhance accuracy in both normal and challenging conditions, particularly for long-tail scenarios. To address this, recent studies have explored two directions of adaptive approaches: MoE-based adaptive fusion, which struggles with uncertainties arising from distinct object configurations, and late fusion for output-level adaptive fusion, which relies on separate detection pipelines and limits comprehensive understanding. In this work, we introduce Cocoon, an object- and feature-level uncertainty-aware fusion framework. The key innovation lies in uncertainty quantification for heterogeneous representations, enabling fair comparison across modalities through the introduction of a feature aligner and a learnable surrogate ground truth, termed feature impression. We also define a training objective to ensure that their relationship provides a valid metric for uncertainty quantification. Cocoon consistently outperforms existing static and adaptive methods in both normal and challenging conditions, including those with natural and artificial corruptions. Furthermore, we show the validity and efficacy of our uncertainty metric across diverse datasets.

Large Language Models (LLMs) are increasingly deployed in large-scale online services, enabling sophisticated applications. However, the computational overhead of generating key-value (KV) caches in the prefill stage presents a major bottleneck, particularly for long-context inputs. Prefix caching mitigates this issue by storing KV caches for reuse, reducing redundant computation. Despite its advantages, prefix caching suffers from high latency due to the limited I/O bandwidth of storage devices, constraining inference efficiency. To address this challenge, we introduce Cake, a novel KV cache loading system that optimally utilizes both computational and I/O resources in parallel. Cake employs a bidirectional scheduling strategy that dynamically balances KV cache computation and loading, ensuring efficient resource utilization. Additionally, Cake incorporates an adaptive scheduling mechanism that seamlessly integrates with non-prefix caching requests, improving system throughput and adapting to fluctuating resource availabilty. Through extensive evaluations across various hardware configurations, datasets, and storage conditions, Cake achieves on average 2.6× reduction in Time to First Token (TTFT) compared to compute-only and I/O-only methods. Our findings highlight Cake as an effective and practical solution for optimizing long-context LLM inference, bridging the gap between computation and I/O efficiency in large-scale AI deployments.

Perception is crucial in the realm of autonomous driving systems, where bird's eye view (BEV)-based architectures have recently reached state-of-the-art performance. The desirability of self-supervised representation learning stems from the expensive and laborious process of annotating 2D and 3D data. Although previous research has investigated pretraining methods for both LiDAR and camera-based 3D object detection, a unified pretraining framework for multimodal BEV perception is missing. In this study, we introduce CALICO, a novel framework that applies contrastive objectives to both LiDAR and camera backbones. Specifically, CALICO incorporates two stages: point-region contrast (PRC) and region-aware distillation (RAD). PRC better balances the region- and scene-level representation learning on the LiDAR modality and offers significant performance improvement compared to existing methods. RAD effectively achieves contrastive distillation on our self-trained teacher model. CALICO's efficacy is substantiated by extensive evaluations on 3D object detection and BEV map segmentation tasks, where it delivers significant performance improvements. Notably, CALICO outperforms the baseline method by 10.5\% and 8.6\% on NDS and mAP. Moreover, CALICO boosts the robustness of multimodal 3D object detection against adversarial attacks and corruption. Additionally, our framework can be tailored to different backbones and heads, positioning it as a promising approach for multimodal BEV perception.

The increased adoption of smart contracts in many industries has made them an attractive target for cybercriminals, leading to millions of dollars in losses. Thus, continuously fixing newly found vulnerabilities of smart contracts becomes a routine software maintenance task for running smart contracts. However, fixing the vulnerabilities that are specific to the smart contract domain requires security knowledge that many developers lack. Without effective tool support, this task can be very costly in terms of manual labor. To fill this critical need, in this paper, we propose VFIX, which automatically generates security patches for vulnerable smart contracts. In particular, VFIX provides a novel program analysis framework that can incorporate different fix patterns for fixing various types of vulnerabilities. To address the unique challenges in accurately fixing smart contract vulnerabilities, VFIX innovatively combines template-based repair with a set of static program analysis techniques specially designed for smart contracts. Specifically, given an input smart contract, VFIX conducts ensemble identification based on multiple static verification tools to identify vulnerabilities for an automatic fix. Then, VFIX generates patches using template-based fix patterns, and conducts static program analysis (e.g., program dependency computation, pointer analysis) for smart contracts to accurately infer and populate the parameter values for the fix templates. Finally, VFIX performs static verification to ensure that the patched contract is free of vulnerabilities. Our evaluations on 144 real smart contracts containing different types of vulnerabilities show that VFIX can successfully fix 94% of the vulnerabilities and preserve the expected normal behaviors of the smart contracts.

Cooperative perception significantly enhances the perception performance of connected autonomous vehicles. Instead of purely relying on local sensors with limited range, it enables multiple vehicles and roadside infrastructures to share sensor data to perceive the environment collaboratively. Through our study, we realize that the performance of cooperative perception systems is limited in real-world deployment due to (1) out-of-sync sensor data during data fusion and (2) inaccurate localization of occluded areas. To address these challenges, we develop RAO, an innovative, effective, and lightweight cooperative perception system that merges asynchronous sensor data from different vehicles through our novel designs of motion-compensated occupancy flow prediction and on-demand data sharing, improving both the accuracy and coverage of the perception system. Our extensive evaluation, including real-world and emulation-based experiments, demonstrates that RAO outperforms state-of-the-art solutions by more than 34% in perception coverage and by up to 14% in perception accuracy, especially when asynchronous sensor data is present. RAO consistently performs well across a wide variety of map topologies and driving scenarios. RAO incurs negligible additional latency (8.5 ms) and low data transmission overhead (10.9 KB per frame), making cooperative perception feasible.

Automotive Ethernet is considered to be the next-generation in-vehicle network, because of its high bandwidth, high throughput, and low cost characteristics. However, no common standard has been established for the security protocol of Automotive Ethernet. While there are a few candidates, including MACsec, IPsec, and TLS, there is no widely favored candidate. Most importantly, existing candidates cannot fully satisfy the requirements of in-vehicle communication, specifically source authentication for broadcast/multicast communication. In this paper, we conduct a comprehensive analysis in both security and performance of existing security protocol candidates and identify source authentication and Denial-of-Service (DoS) prevention as two essential but missing properties in these candidates. We propose Gatekeeper, a gateway-based broadcast authentication protocol to ensure source authentication. In general, Gatekeeper introduces an on-path authenticator, which co-locates with the in-vehicle gateway or domain controllers and helps receivers to verify the sender's identity. To defend against DoS threats, we further integrate the time-lock puzzle with Gatekeeper to slow down malicious traffic. Our performance evaluation results show that Gatekeeper only results in 0.03 ms latency overhead for CAN data transmission and outperforms TESLA on both CAN and LiDAR transmission scenarios, highlighting the effectiveness and efficiency of Gatekeeper.

Trajectory prediction is a critical component for autonomous vehicles (AVs) to perform safe planning and navigation. However, few studies have analyzed the adversarial robustness of trajectory prediction or investigated whether the worst-case prediction can still lead to safe planning. To bridge this gap, we study the adversarial robustness of trajectory prediction models by proposing a new adversarial attack that perturbs normal vehicle trajectories to maximize the prediction error. Our experiments on three models and three datasets show that the adversarial prediction increases the prediction error by more than 150%. Our case studies show that if an adversary drives a vehicle close to the target AV following the adversarial trajectory, the AV may make an inaccurate prediction and even make unsafe driving decisions. We also explore possible mitigation techniques via data augmentation and trajectory smoothing.

Autonomous vehicles (AVs) are on the verge of changing the transportation industry. Despite the fast development of autonomous driving systems (ADSs), they still face safety and security challenges. Current defensive approaches usually focus on a narrow objective and are bound to specific platforms, making them difficult to generalize. To solve these limitations, we propose AVMaestro, an efficient and effective policy enforcement framework for full-stack ADSs. AVMaestro includes a code instrumentation module to systematically collect required information across the entire ADS, which will then be feed into a centralized data examination module, where users can utilize the global information to deploy defensive methods to protect AVs from various threats. AVMaestro is evaluated on top of Apollo-6.0 and experimental results confirm that it can be easily incorporated into the original ADS with almost negligible run-time delay. We further demonstrate that utilizing the global information can not only improve the accuracy of existing intrusion detection methods, but also potentially inspire new security applications.

Cyber-physical industrial control systems (ICS) commonly implement configuration parameters that can be remotely tuned by human-machine interfaces (HMI) at runtime. These parameters directly control the behaviors of ICSs thus they can be exploited by attackers to compromise the safety of ICSs, proved by real-world attacks worldwide. However, existing anomaly detection methods, which mostly focus on the programmable logic controller (PLC) programs or sensor signals, lack a comprehensive analysis of configuration’s impact on the entire system and thus cannot effectively detect improper parameters. A tool that automatically analyzes complicated control logic to determine the safety of configuration is absent. To fill this gap, we design SmtConf, a verification-based framework for detecting and mitigating improper parameters in ICSs at runtime. To understand the impact of configuration parameters on complicated control logic, we design a symbolic formal model representing behaviors of the ICS under any possible configuration parameters. Based on the model, SmtConf works as a monitoring system that detects safety violations in real-time when the improper configuration is injected. To further assist developers to determine the safe configuration, SmtConf recommends safe configuration parameters by solving an optimization problem. In 18 test cases collected from two production-level ICS testbeds, SmtConf detects all true violations caused by improper parameters in 0.41 seconds and correctly repairs the ICS with recommended safe parameters in 0.45 seconds.

Ring confidential transaction (RingCT) protocol is widely used in cryptocurrency to protect the privacy of both users’ identities and transaction amounts. Most recently, a new RingCT protocol (called RingCT 2.0) was proposed by leveraging cryptographic accumulators, which can achieve a constant-size output theoretically but still far from being practical due to the heavy zero-knowledge associated with the accumulator. In this article, we revisit the design of ring confidential transaction protocol and put forward a more efficient privacy-preserving payment protocol, which is built upon an extended version of one-out-of-many proof and a special multi-signature. Compared with previous works, the new protocol is not only more practical, but also does not suffer from a trusted setup. Besides, we show that the protocol satisfies the security requirements provided that the underlying cryptographic primitives are secure in the random oracle model. We implement our new payment protocol in Java, and the experimental results show that it is efficient enough to be used in practice.

Smart contracts, programs running on blockchain systems, leverage diverse decentralized applications (DApps). Unfortunately, well-known smart contract platforms, Ethereum for example, face serious security problems. Exploits to contracts may cause enormous financial losses, which emphasize the importance of smart contract testing. However, current exploit generation tools have difficulty to solve hard constraints in execution paths and cannot simulate the blockchain behaviors very well. These problems cause a loss of coverage and accuracy of exploit generation. To overcome the problems, we design and implement EthPloit, a smart contract exploit generator based on fuzzing. EthPloit adopts static taint analysis to generate exploit-targeted transaction sequences, a dynamic seed strategy to pass hard constraints and an instrumented Ethereum Virtual Machine to simulate blockchain behaviors. We evaluate EthPloit on 45,308 smart contracts and discovered 554 exploitable contracts. EthPloit automatically generated 644 exploits without any false positive and 306 of them cannot be generated by previous exploit generation tools.