Test out our new Bento Search

No summary available.

No summary available.

Recently, object detection has proven vulnerable to adversarial patch attacks. The attackers holding a specially crafted patch can hide themselves from state-of-the-art detectors, e.g., YOLO, even in the physical world. This attack can bring serious security threats, such as escaping from surveillance cameras. How to effectively detect this kind of adversarial examples to catch potential attacks has become an important problem. In this paper, we propose two detection methods: the signature-based method and the signature-independent method. First, we identify two signatures of existing adversarial patches that can be utilized to precisely locate patches within adversarial examples. By employing the signatures, a fast signature-based method is developed to detect the adversarial objects. Second, we present a robust signature-independent method based on the content semantics consistency of model outputs. Adversarial objects violate this consistency, appearing locally but disappearing globally, while benign ones remain consistently present. The experiments demonstrate that two proposed methods can effectively detect attacks both in the digital and physical world. These methods each offer distinct advantage. Specifically, the signature-based method is capable of real-time detection, while the signature-independent method can detect unknown adversarial patch attacks and makes defense-aware attacks almost impossible to perform.

The reduction of CO2 into value-added chemicals offers a promising approach to mitigate air pollution while simultaneously generating economic value. In this context, the chemical fixation of CO2 into epoxides to generate cyclic carbonates is a sustainable technique due to its high atom efficiency. In this work, we report the preparation of simple iron-based ionic liquids (ILs) derived from hexacyanoferrate­(III), (Fe­(CN)6), which exhibit remarkable activity and selectivity toward cyclic carbonate formation. Molecular dynamics (MD) simulations demonstrate that the contact ion pair organization in the IL is anisotropic, exhibiting a distinct spatial arrangement. The IL efficiently catalyzed the conversion of various epoxides using only 1.0 mol % IL under mild conditions (1–2 bar, 70–100 °C). Moreover, solid catalysts containing ionic liquid layers (SCILLs), akin to catch-and-release catalytic systems, are developed that demonstrate remarkable activity, achieving turnover numbers (TONs) of 265–729 for aliphatic epoxides and 83–668 for aromatic epoxides, with 99% selectivity toward cyclic carbonates under the same mild conditions. A monolayer of IL enhances local charge density by aligning cations and anions into distinct layers on SiO2, therefore creating nanoconfined spaces within the SCILL (solid catalysts with IL layer). These confined domains function as a “catch-and-release” catalytic system, controlling the diffusion of epoxides, CO2, and intermediates toward the active sites while facilitating the release of products from the microionic environment. An in situ NMR study conducted under realistic experimental conditions revealed that the reaction mechanism involves the formation of 1-n-butyl-3-methylimidazolium-2-carboxylate (NHC–CO2) intermediate, thereby challenging the classical understanding of IL-assisted catalysis and providing new fundamental insights into the field.

The impact of recreational fisheries on marine ecosystems is often overshadowed by commercial fisheries, although recreational fishing harvest can be substantial, especially for species that are either overfished or experiencing overfishing. Delaware is a small coastal state with ~1,000,000 residents and nearly 272,000 resident and non‐resident anglers. We used publicly available data for Delaware's recreational fisheries during 1981–2021 to determine the nine most caught fish species and to evaluate trends in total numbers caught, harvested, released, and catch per unit effort (CPUE). The top nine most frequently captured fish by recreational anglers were Summer Flounder (Paralichthys dentatus), Atlantic Croaker (Micropogonias undulatus), Bluefish (Pomatomus saltatrix), Black Sea Bass (Centropristis striata), Weakfish (Cynoscion regalis), White Perch (Morone americana), Tautog (Tautoga onitis), Striped Bass (Morone saxatilis), and Spot (Leiostomus xanthurus). The proportion of fish released increased through time for all nine species, suggesting that the recreational fishery in Delaware is transitioning from a harvest‐oriented to a catch‐and‐release‐oriented fishery. Observations of higher release rates in recreational fisheries of Delaware are consistent with the findings elsewhere in the world for freshwater and marine systems. [ABSTRACT FROM AUTHOR]

The rapid development of Artificial Intelligence (AI) technology must be connected to the arithmetic support of high-performance hardware. However, when the deep neural network (DNN) accelerator performs inference tasks at the edge end, the sensitive data of DNN will generate leakage through side-channel information. The adversary can recover the model structure and weight parameters of DNN by using the side-channel information, which seriously affects the protection of necessary intellectual property (IP) of DNN, so the hardware security of the DNN accelerator is critical. In the current research of Side-channel attack (SCA) for matrix multiplication units, such as systolic arrays, the linear multiplication operation leads to a more extensive weights search space for the SCA, and extracting all the weight parameters requires higher attack conditions. This article proposes a new power SCA method, which includes a Collision-Correlation Power Analysis (Collision-CPA) and Correlation-based Weight Search Algorithm (C-WSA) to address the problem. The Collision-CPA reduces the attack conditions for the SCA by building multiple Hamming Distance (HD)-based power leakage models for the systolic array. Meanwhile, the C-WSA dramatically reduces the weights search space. In addition, the concept of a Side-channel star map (SCSM) is proposed for the first time in this article, and the adversary can quickly and accurately locate the correct weight information in the SCSM. Through experiments, we recover all the weight parameters of a $3\times 3$ systolic array based on 100000 power traces, in which the weight search space is reduced by up to 97.7%. For the DNN accelerator at the edge, especially the systolic array structure, our proposed novel SCA aligns more with practical attack scenarios, with lower attack conditions, and higher attack efficiency.