作者：
Kan Zhang;Mugang Lin*;Lingzhi Zhu;Yunhui Wang;Wenzhuo He
期刊：
Biomedical Signal Processing and Control,2026年113:108868 ISSN：1746-8094
通讯作者：
Mugang Lin
作者机构：
[Kan Zhang; Yunhui Wang; Wenzhuo He] Hengyang Normal University of College of Computer Science and Technology, Hengyang, 421002, Hunan, China;[Lingzhi Zhu] School of Computer Science and Engineering, Hunan Institute of Technology, Hengyang, 421002, Hunan, China;Hunan Engineering Research Center of Cyberspace Security Technology and Applications, Hengyang, 421002, Hunan, China;Hunan Provincial Key Laboratory of Intelligent Information Processing and Application, Hengyang, 421002, Hunan, China;[Mugang Lin] Hengyang Normal University of College of Computer Science and Technology, Hengyang, 421002, Hunan, China<&wdkj&>Hunan Engineering Research Center of Cyberspace Security Technology and Applications, Hengyang, 421002, Hunan, China<&wdkj&>Hunan Provincial Key Laboratory of Intelligent Information Processing and Application, Hengyang, 421002, Hunan, China
通讯机构：
[Mugang Lin] H;Hengyang Normal University of College of Computer Science and Technology, Hengyang, 421002, Hunan, China<&wdkj&>Hunan Engineering Research Center of Cyberspace Security Technology and Applications, Hengyang, 421002, Hunan, China<&wdkj&>Hunan Provincial Key Laboratory of Intelligent Information Processing and Application, Hengyang, 421002, Hunan, China
摘要：
Gene regulatory networks (GRNs) have revealed the internal mechanism and complex relationship of gene expression regulation, and its research is of great significance for the in-depth understanding of life activities and accurate disease diagnosis. Although the existing methods can realize the expression analysis at the cell level with the promotion of single-cell RNA sequencing(scRNA-seq) technology, most focus on the interaction between local genes, and it is difficult to capture the overall organizational structure and long-term regulatory effects. In this study, we present a novel supervised method named KA4GANC, which integrates the Kolmogorov–Arnold Network (KAN) with a Graph Attention Network (GAT) to address the critical limitations in capturing global regulatory architecture from scRNA-seq data. KA4GANC’s novelty lies in two key components. First, it leverages a Fourier KAN for nonlinear feature transformation via adaptive, multi-scale Fourier basis functions, thereby generating highly expressive gene embeddings. Second, it replaces linear transformations in graph attention layers with a KAN-based convolution, enabling the model to learn complex nonlinear local interactions and effectively preserve neighborhood topology in the latent space. Benchmark evaluations on seven scRNA-seq datasets across three ground-truth network types demonstrate KA4GANC’s state-of-the-art performance, achieving average AUROC of 0.84 with 34.6% faster convergence.
Gene regulatory networks (GRNs) have revealed the internal mechanism and complex relationship of gene expression regulation, and its research is of great significance for the in-depth understanding of life activities and accurate disease diagnosis. Although the existing methods can realize the expression analysis at the cell level with the promotion of single-cell RNA sequencing(scRNA-seq) technology, most focus on the interaction between local genes, and it is difficult to capture the overall organizational structure and long-term regulatory effects. In this study, we present a novel supervised method named KA4GANC, which integrates the Kolmogorov–Arnold Network (KAN) with a Graph Attention Network (GAT) to address the critical limitations in capturing global regulatory architecture from scRNA-seq data. KA4GANC’s novelty lies in two key components. First, it leverages a Fourier KAN for nonlinear feature transformation via adaptive, multi-scale Fourier basis functions, thereby generating highly expressive gene embeddings. Second, it replaces linear transformations in graph attention layers with a KAN-based convolution, enabling the model to learn complex nonlinear local interactions and effectively preserve neighborhood topology in the latent space. Benchmark evaluations on seven scRNA-seq datasets across three ground-truth network types demonstrate KA4GANC’s state-of-the-art performance, achieving average AUROC of 0.84 with 34.6% faster convergence.
作者：
Peng Tang;Huihuang Zhao*;Weiliang Meng;Yaonan Wang
期刊：
Expert Systems with Applications,2026年297:129344 ISSN：0957-4174
通讯作者：
Huihuang Zhao
作者机构：
[Peng Tang; Huihuang Zhao] Hengyang Normal University, College of Computer Science and Technology, Hengyang, 421002, Hunan, China;[Weiliang Meng] University of Chinese Academy of Sciences, School of Artificial Intelligence, Beijing, 100049, China;[Yaonan Wang] Hunan University, National Engineering Laboratory for Robot Visual Perception and Control Technology, Changsha, 410082, Hunan, China
通讯机构：
[Huihuang Zhao] H;Hengyang Normal University, College of Computer Science and Technology, Hengyang, 421002, Hunan, China
摘要：
Exciting achievements have been made in audio-driven face-talking head generation. However, while existing methods excel at generating a speaking head from a frontal identity image, generating a speaking head with a head pose does not yield satisfactory results when the identity is based on a side image of the face. To address this limitation, a concise and effective approach is proposed in this work. Our method generates efficient talking head videos using a side face image as the identity. It uses facial features and head posture to predict frontal keypoints, during which facial expression features are added as additional features. The method achieves impressive head motion effects and maintains identity information consistency during motion. Additionally, a Transform-based lip-sync expert discriminator is designed to guide the lip synchronization of the generated video. A cross-attention is employed to jointly learn the front and back frames, extracting contextual information that improves lip synchronization over long sequences. Our method is tested on several datasets, such as VoxCeleb2 and LRS2, and compared with multiple advanced methods, demonstrating its outstanding performance through experiments.
Exciting achievements have been made in audio-driven face-talking head generation. However, while existing methods excel at generating a speaking head from a frontal identity image, generating a speaking head with a head pose does not yield satisfactory results when the identity is based on a side image of the face. To address this limitation, a concise and effective approach is proposed in this work. Our method generates efficient talking head videos using a side face image as the identity. It uses facial features and head posture to predict frontal keypoints, during which facial expression features are added as additional features. The method achieves impressive head motion effects and maintains identity information consistency during motion. Additionally, a Transform-based lip-sync expert discriminator is designed to guide the lip synchronization of the generated video. A cross-attention is employed to jointly learn the front and back frames, extracting contextual information that improves lip synchronization over long sequences. Our method is tested on several datasets, such as VoxCeleb2 and LRS2, and compared with multiple advanced methods, demonstrating its outstanding performance through experiments.
期刊：
SPECTROCHIMICA ACTA PART A-MOLECULAR AND BIOMOLECULAR SPECTROSCOPY,2026年346:126921 ISSN：1386-1425
通讯作者：
Li, Zhongliang;Chen, Wen
作者机构：
[Liu, Mengqin; Chen, Taiyi; Lin, Xiaoping; Zeng, Rongying; Jiang, Lu; Li, Peisong; Huang, Leyuan] Hunan Engineering Research Center for Monitoring and Treatment of Heavy Metal Pollution in the Upper Reaches of Xiangjiang River, College of Chemistry and Material Science, Hengyang Normal University, Hengyang 421001, PR China;[Liu, Wei] Hunan Hengyang Ecological Environment Monitoring Center, Hengyang, Hunan 421001, China;[Li, Zhongliang] Hunan Engineering Research Center for Monitoring and Treatment of Heavy Metal Pollution in the Upper Reaches of Xiangjiang River, College of Chemistry and Material Science, Hengyang Normal University, Hengyang 421001, PR China. Electronic address: zhongliang.li@hynu.edu.cn;[Zhu, Ruirui] Hunan Provincial Ecological and Environmental Monitoring Center, State Environmental Protection Key Laboratory of Monitoring for Heavy Metal Pollution, Changsha 410019, China;[Chen, Zhi-Liang] School of Pharmacy, Shaoyang University, Shaoyang 422000, PR China
通讯机构：
[Chen, Wen; Li, Zhongliang] H;Hunan Engineering Research Center for Monitoring and Treatment of Heavy Metal Pollution in the Upper Reaches of Xiangjiang River, College of Chemistry and Material Science, Hengyang Normal University, Hengyang 421001, PR China. Electronic address:
关键词：
AIE;J-aggregate;Ratiometric fluorescence sensors;Zinc and copper ions
摘要：
The precise detection of metal ions is crucial in food safety, biomedical diagnostics, and environmental monitoring. Although traditional “on-off” fluorescence probes are frequently utilized, their anti-interference capability remains a limitation. This study synthesized three new copper-based fluorescent probes: PTA, PTP, and PTD. Among them, PTD exhibited unique aggregation-induced emission (AIE) properties. PTD's remarkable selectivity for Zn 2+ was due to the formation of Zn 2+ -induced J-aggregates, which resulted in a concentration-dependent fluorescence redshift from green to yellow and then to red. Additionally, PTD demonstrated a rapid and sensitive response to Cu 2+ , and it was successfully integrated into a smartphone-based platform for the semi-quantitative detection of Zn 2+ and Cu 2+ in real food samples. The effectiveness in biological applications was also validated through live-cell imaging tests. This work contributes to the creation of highly selective and interference-resistant ratiometric fluorescence sensors by leveraging the special process of target metal ion-induced J-aggregate production in AIE-based probes.
The precise detection of metal ions is crucial in food safety, biomedical diagnostics, and environmental monitoring. Although traditional “on-off” fluorescence probes are frequently utilized, their anti-interference capability remains a limitation. This study synthesized three new copper-based fluorescent probes: PTA, PTP, and PTD. Among them, PTD exhibited unique aggregation-induced emission (AIE) properties. PTD's remarkable selectivity for Zn 2+ was due to the formation of Zn 2+ -induced J-aggregates, which resulted in a concentration-dependent fluorescence redshift from green to yellow and then to red. Additionally, PTD demonstrated a rapid and sensitive response to Cu 2+ , and it was successfully integrated into a smartphone-based platform for the semi-quantitative detection of Zn 2+ and Cu 2+ in real food samples. The effectiveness in biological applications was also validated through live-cell imaging tests. This work contributes to the creation of highly selective and interference-resistant ratiometric fluorescence sensors by leveraging the special process of target metal ion-induced J-aggregate production in AIE-based probes.
摘要：
Radon (Rn-222) released by the decay of Ra-226 in uranium ores may cause potential radiation risks in the water environment. However, the migration behaviour characterised by the low diffusion coefficient of radon in water and the slow change in concentration is still lacking systematic research. This study investigates the radon release behavior of uranium ore in water under different temperature conditions (20–32 °C). A closed-loop measurement system combined with a bubbling method was established, utilizing a RAD7 radon detector. A nonlinear model was developed to describe the time-dependent measured radon concentration, and the radon release rate was obtained through nonlinear fitting. Experiments were conducted under five temperature conditions ranging from 20 °C to 32 °C. The Results showed that the proposed model achieved high fitting accuracy (R 2 > 0.95) across all conditions. The radon release rate increased significantly with temperature, rising from 0.210 ± 0.006 mBq s −1 to 0.391 ± 0.019 mBq s −1 within the range of 20 °C–32 °C, an increase of 86 %. Further analysis shows that the radon release rate of uranium ore in water has a good linear relationship with temperature, indicating that temperature is an important factor affecting the radon release behavior. The methods and models proposed in this paper can effectively depict the radon release behavior of uranium ores in water environments and are suitable for related experimental designs to assess environmental radiation risks.
Radon (Rn-222) released by the decay of Ra-226 in uranium ores may cause potential radiation risks in the water environment. However, the migration behaviour characterised by the low diffusion coefficient of radon in water and the slow change in concentration is still lacking systematic research. This study investigates the radon release behavior of uranium ore in water under different temperature conditions (20–32 °C). A closed-loop measurement system combined with a bubbling method was established, utilizing a RAD7 radon detector. A nonlinear model was developed to describe the time-dependent measured radon concentration, and the radon release rate was obtained through nonlinear fitting. Experiments were conducted under five temperature conditions ranging from 20 °C to 32 °C. The Results showed that the proposed model achieved high fitting accuracy (R 2 > 0.95) across all conditions. The radon release rate increased significantly with temperature, rising from 0.210 ± 0.006 mBq s −1 to 0.391 ± 0.019 mBq s −1 within the range of 20 °C–32 °C, an increase of 86 %. Further analysis shows that the radon release rate of uranium ore in water has a good linear relationship with temperature, indicating that temperature is an important factor affecting the radon release behavior. The methods and models proposed in this paper can effectively depict the radon release behavior of uranium ores in water environments and are suitable for related experimental designs to assess environmental radiation risks.
作者机构：
College of Computer Science and Technology, Hengyang Normal University, Hengyang, 421002, China;Hunan Provincial Key Laboratory of Intelligent Information Processing and Application, Hengyang Normal University, Hengyang, 421002, China;Hunan Engineering Research Center of Cyberspace Security Technology and Applications, Hengyang Normal University, Hengyang, 421002, China;[Xingqi Yue; Qingling Song] College of Computer Science and Technology, Hengyang Normal University, Hengyang, 421002, China<&wdkj&>Hunan Provincial Key Laboratory of Intelligent Information Processing and Application, Hengyang Normal University, Hengyang, 421002, China<&wdkj&>Hunan Engineering Research Center of Cyberspace Security Technology and Applications, Hengyang Normal University, Hengyang, 421002, China
摘要：
Lightweight cryptography is crucial for securing data in resource-constrained IoT devices. However, most existing lightweight block ciphers suffer from slow diffusion, high hardware cost, or insufficient side-channel resistance, limiting their practical deployment. To address these challenges, this work proposes HDHL, a 64-bit block cipher that integrates a hybrid Generalized Feistel and Substitution-Permutation Network (GSP) structure. The design features a 64-bit sponge-based F -function for strong nonlinearity, lightweight AND-Rotation-XOR (AND-RX) operations for enhanced confusion, and an involutive linear layer that enables efficient encryption and decryption with a unified datapath. A compact LFSR-driven key schedule delivers four 16-bit sub-keys per round with near-ideal entropy. Experimental results show that HDHL achieves full diffusion in just two rounds, area occupies only 1756.71 gate equivalents (GE), and energy 7.82 μ J/bit at a 100 kHz test clock, outperforming CRAFT and SKINNY-64 in both area and energy. Security evaluation demonstrates strong resistance against differential, linear, integral, and algebraic attacks, with a seven-round differential probability below 2 − 70 . Fixed-random t -tests on a prototype ASIC confirm first-order side-channel resistance when lightweight masking or a two-share threshold implementation is applied. These results indicate that HDHL offers a balanced combination of security, efficiency, and implementation cost, making it a promising candidate for future low-power embedded applications.
Lightweight cryptography is crucial for securing data in resource-constrained IoT devices. However, most existing lightweight block ciphers suffer from slow diffusion, high hardware cost, or insufficient side-channel resistance, limiting their practical deployment. To address these challenges, this work proposes HDHL, a 64-bit block cipher that integrates a hybrid Generalized Feistel and Substitution-Permutation Network (GSP) structure. The design features a 64-bit sponge-based F -function for strong nonlinearity, lightweight AND-Rotation-XOR (AND-RX) operations for enhanced confusion, and an involutive linear layer that enables efficient encryption and decryption with a unified datapath. A compact LFSR-driven key schedule delivers four 16-bit sub-keys per round with near-ideal entropy. Experimental results show that HDHL achieves full diffusion in just two rounds, area occupies only 1756.71 gate equivalents (GE), and energy 7.82 μ J/bit at a 100 kHz test clock, outperforming CRAFT and SKINNY-64 in both area and energy. Security evaluation demonstrates strong resistance against differential, linear, integral, and algebraic attacks, with a seven-round differential probability below 2 − 70 . Fixed-random t -tests on a prototype ASIC confirm first-order side-channel resistance when lightweight masking or a two-share threshold implementation is applied. These results indicate that HDHL offers a balanced combination of security, efficiency, and implementation cost, making it a promising candidate for future low-power embedded applications.
期刊：
Journal of the Taiwan Institute of Chemical Engineers,2026年180:106480 ISSN：1876-1070
通讯作者：
Wenjin Ni
作者机构：
[Shuilin Liu; Ning Liu; Xinde Tang; Suyun Wu; Aiming Sun; Aiyang Li] School of chemical and environmental engineering, Hunan Institute of Technology, Hengyang 421002, China;College of Chemistry and Materials Science, Hengyang Normal University, Hengyang 421008, China;Key Laboratory of Functional Metal-Organic Compounds of Hunan Province, Hengyang 421008, China;[Wenjin Ni] College of Chemistry and Materials Science, Hengyang Normal University, Hengyang 421008, China<&wdkj&>Key Laboratory of Functional Metal-Organic Compounds of Hunan Province, Hengyang 421008, China
通讯机构：
[Wenjin Ni] C;College of Chemistry and Materials Science, Hengyang Normal University, Hengyang 421008, China<&wdkj&>Key Laboratory of Functional Metal-Organic Compounds of Hunan Province, Hengyang 421008, China
摘要：
Background The growing demand for ε-caprolactam, a key precursor for nylon-6 production, has driven strategic interest in developing cost-effective and environmentally benign approaches for cyclohexanone oxime synthesis.
The growing demand for ε-caprolactam, a key precursor for nylon-6 production, has driven strategic interest in developing cost-effective and environmentally benign approaches for cyclohexanone oxime synthesis.
Methods Ti-incorporated hexagonal mesoporous silica (TiO 2 /Ti-HMS) catalysts with tunable Ti/Si molar ratios were in-situ synthesized, and applied in the selective aerobic oxidation of cyclohexylamine to cyclohexanone oxime. The textural properties of TiO 2 /Ti-HMS were characterized, and the bifunctional effects of surface hydroxyls sites and Brønsted acid centers were investigated. The adsorption energies of both cyclohexylamine and nitrosocyclohexane intermediate on different catalysts were calculated by density functional theory (DFT).
Ti-incorporated hexagonal mesoporous silica (TiO 2 /Ti-HMS) catalysts with tunable Ti/Si molar ratios were in-situ synthesized, and applied in the selective aerobic oxidation of cyclohexylamine to cyclohexanone oxime. The textural properties of TiO 2 /Ti-HMS were characterized, and the bifunctional effects of surface hydroxyls sites and Brønsted acid centers were investigated. The adsorption energies of both cyclohexylamine and nitrosocyclohexane intermediate on different catalysts were calculated by density functional theory (DFT).
Significant Findings The incorporation of partial titanium species into the HMS framework via Ti-O-Si bonds, while the remaining titanium formed highly dispersed TiO 2 nanoparticles to form the TiO 2 /Ti-HMS catalyst. The surface hydroxyls on anatase TiO 2 nanoparticles promoted the adsorption and activation of cyclohexylamine. Meanwhile, the skeletal titanium in the HMS framework enriched the Brønsted acid centers, which promoted the rearrangement of the nitrosocyclohexane intermediate to cyclohexanone oxime. Bifunctional Ti-HMS-0.05 exhibited superior catalytic performance and stability. A possible reaction pathway was proposed based on the experimental results and DFT calculations.
The incorporation of partial titanium species into the HMS framework via Ti-O-Si bonds, while the remaining titanium formed highly dispersed TiO 2 nanoparticles to form the TiO 2 /Ti-HMS catalyst. The surface hydroxyls on anatase TiO 2 nanoparticles promoted the adsorption and activation of cyclohexylamine. Meanwhile, the skeletal titanium in the HMS framework enriched the Brønsted acid centers, which promoted the rearrangement of the nitrosocyclohexane intermediate to cyclohexanone oxime. Bifunctional Ti-HMS-0.05 exhibited superior catalytic performance and stability. A possible reaction pathway was proposed based on the experimental results and DFT calculations.
作者机构：
State Key Laboratory of Silicate Materials for Architectures, Wuhan University of Technology, Wuhan 430070, China;[Quanyao Zhu] School of Materials Science and Engineering, Wuhan University of Technology, Wuhan 430070 Hubei, China;Advanced Ceramics Institute of Zibo New & High-Tech Industrial Development Zone, Zibo 255000 Shandong, China;College of Chemistry and Materials Science, Hengyang Normal University, Hengyang 421002, China;[Jianing Zhu; Jia He; Piyu Gong; Fuling Wu; Wenqi Sun; Zijin Meng; Qingqing Zhang; Dongchun Qiu; Huajun Sun] State Key Laboratory of Silicate Materials for Architectures, Wuhan University of Technology, Wuhan 430070, China<&wdkj&>School of Materials Science and Engineering, Wuhan University of Technology, Wuhan 430070 Hubei, China<&wdkj&>Advanced Ceramics Institute of Zibo New & High-Tech Industrial Development Zone, Zibo 255000 Shandong, China
通讯机构：
[Quanyao Zhu] S;School of Materials Science and Engineering, Wuhan University of Technology, Wuhan 430070 Hubei, China
摘要：
This paper demonstrates a heterostructured capacitor combining low-hysteresis Pb 0.94 La 0.06 ZrO 3 (PLZ) with high-polarization Pb 0.94 La 0.06 (Zr 0.52 Ti 0.48 )O 3 (PLZT) on LaNiO3/Si substrates. Strategic insertion of Al 2 O 3 (AO) interlayer at PLZT/PLZ interface effectively suppressed leakage currents while optimizing electric field distribution, substantially enhancing breakdown field strength from 2057 kV/cm to 3620 kV/cm. Optimized PLZT/AO/PLZ heterostructure exhibited outstanding energy storage performance, achieving record-high recoverable energy density of 101.4 J/cm 3 with 74.7 % energy conversion efficiency, providing a breakthrough solution for lead zirconate titanate-based thin film energy storage devices.
This paper demonstrates a heterostructured capacitor combining low-hysteresis Pb 0.94 La 0.06 ZrO 3 (PLZ) with high-polarization Pb 0.94 La 0.06 (Zr 0.52 Ti 0.48 )O 3 (PLZT) on LaNiO3/Si substrates. Strategic insertion of Al 2 O 3 (AO) interlayer at PLZT/PLZ interface effectively suppressed leakage currents while optimizing electric field distribution, substantially enhancing breakdown field strength from 2057 kV/cm to 3620 kV/cm. Optimized PLZT/AO/PLZ heterostructure exhibited outstanding energy storage performance, achieving record-high recoverable energy density of 101.4 J/cm 3 with 74.7 % energy conversion efficiency, providing a breakthrough solution for lead zirconate titanate-based thin film energy storage devices.
期刊：
Personality and Individual Differences,2026年248:113492 ISSN：0191-8869
通讯作者：
Ya Zheng
作者机构：
[Lihua Zhou] School of Educational Science, Hengyang Normal University, Hengyang, China;[Hailing Wen] Peking University Sixth Hospital, Peking University Institute of Mental Health, Beijing, China;Department of Psychology, Guangzhou University, Guangzhou, China;Center for Reward and Social Cognition, School of Education, Guangzhou University, Guangzhou, China;[Shiting Chen; Guanglong Liu; Ya Zheng] Department of Psychology, Guangzhou University, Guangzhou, China<&wdkj&>Center for Reward and Social Cognition, School of Education, Guangzhou University, Guangzhou, China
通讯机构：
[Ya Zheng] D;Department of Psychology, Guangzhou University, Guangzhou, China<&wdkj&>Center for Reward and Social Cognition, School of Education, Guangzhou University, Guangzhou, China
摘要：
Sensation seeking strongly predicts diverse risk-taking behaviors, potentially linked to altered neural reward processing. However, it remains unexplored how this reward processing is modulated by uncertainty context. We recorded EEG from 41 high sensation seekers (HSS) and 40 low sensation seekers (LSS) when they anticipated and experienced monetary rewards in risky (known outcome probabilities) and ambiguous (unknown outcome probabilities) contexts. Our results showed that during cue evaluation, HSS exhibited a blunted cue-P3 differentiation between risk and ambiguity compared to LSS. During outcome evaluation, while the reward positivity valence effect was comparable across groups and contexts, HSS exhibited a stronger valence effect on the feedback-P3 under ambiguity than under risk, driven by their undifferentiated feedback-P3 response to gains across uncertainty contexts. These findings offer the first neural evidence that risk-taking vulnerability in sensation seeking may arise from a fundamentally blunted neural distinction between risk and ambiguity in reward dynamics.
Sensation seeking strongly predicts diverse risk-taking behaviors, potentially linked to altered neural reward processing. However, it remains unexplored how this reward processing is modulated by uncertainty context. We recorded EEG from 41 high sensation seekers (HSS) and 40 low sensation seekers (LSS) when they anticipated and experienced monetary rewards in risky (known outcome probabilities) and ambiguous (unknown outcome probabilities) contexts. Our results showed that during cue evaluation, HSS exhibited a blunted cue-P3 differentiation between risk and ambiguity compared to LSS. During outcome evaluation, while the reward positivity valence effect was comparable across groups and contexts, HSS exhibited a stronger valence effect on the feedback-P3 under ambiguity than under risk, driven by their undifferentiated feedback-P3 response to gains across uncertainty contexts. These findings offer the first neural evidence that risk-taking vulnerability in sensation seeking may arise from a fundamentally blunted neural distinction between risk and ambiguity in reward dynamics.
摘要：
Due to its positivity-preserving, entropy condition satisfying and easy extension to other types of hyperbolic equations, the HLLC scheme has become a popular Riemann solver to calculate numerical fluxes. However, there are two drawbacks needed to be tackled before it becomes an impeccable flux solver for various compressible flows: one is the numerical instability in calculating multidimensional strong shock waves; The other is the failure to converge to the desired limit solution in calculating low Mach number flows approaching the incompressible limit. In the current work, the shock instability of the HLLC scheme is cured by simply modifying the nonlinear wave speeds and a hybrid strategy is adopted for accurate calculations of contact waves and rarefaction waves. In addition, the performance of the HLLC scheme in simulating low Mach number flows is improved by controlling the excessive numerical dissipation in momentum equations under low-speed flow regime. The excellent performance of the proposed scheme for simulating flow problems across high and low Mach numbers are demonstrated by a suit of canonical numerical test cases.
Due to its positivity-preserving, entropy condition satisfying and easy extension to other types of hyperbolic equations, the HLLC scheme has become a popular Riemann solver to calculate numerical fluxes. However, there are two drawbacks needed to be tackled before it becomes an impeccable flux solver for various compressible flows: one is the numerical instability in calculating multidimensional strong shock waves; The other is the failure to converge to the desired limit solution in calculating low Mach number flows approaching the incompressible limit. In the current work, the shock instability of the HLLC scheme is cured by simply modifying the nonlinear wave speeds and a hybrid strategy is adopted for accurate calculations of contact waves and rarefaction waves. In addition, the performance of the HLLC scheme in simulating low Mach number flows is improved by controlling the excessive numerical dissipation in momentum equations under low-speed flow regime. The excellent performance of the proposed scheme for simulating flow problems across high and low Mach numbers are demonstrated by a suit of canonical numerical test cases.
作者：
Caizhi Wu;Yipeng Zhao*;Liang Ma;Yicheng Wang;Zhiqiang Li
期刊：
Journal of Physics and Chemistry of Solids,2026年208:113216 ISSN：0022-3697
通讯作者：
Yipeng Zhao
作者机构：
[Caizhi Wu; Yipeng Zhao; Liang Ma; Yicheng Wang; Zhiqiang Li] College of Physics and Electronic Engineering, Hengyang Normal University, Hengyang, 421008, China
通讯机构：
[Yipeng Zhao] C;College of Physics and Electronic Engineering, Hengyang Normal University, Hengyang, 421008, China
摘要：
Two-dimensional van der Waals heterojunctions are seen as a powerful strategy to tune the electronic properties and enhance their performance in devices. In this study, we have systematically investigated the electronic properties and energy band alignments of MoS 2 /MSe (M = In, Ga) heterojunctions using first-principles calculations. The MoS 2 /InSe heterojunction exhibits a type-I band alignment, whereas the MoS 2 /GaSe heterojunction presents a type-II band characteristic. Furthermore, the MoS 2 /InSe heterojunction achieves a transition from type-I to type-II energy band alignment in the strain range of −3 % to −4 % biaxial strain. In contrast, the MoS 2 /GaSe heterojunction consistently maintains the type-II energy band alignment. In addition, the tensile strain effectively enhanced the optical absorption of MoS 2 /GaSe in the visible spectral interval, and a red-shift/blue-shift of the absorption peaks with increasing tensile/compressive strains was observed. The MoS 2 /InSe heterojunction at −3 % and the MoS 2 /GaSe heterojunction at −2 % to −1 % strain intervals exhibited suitable band gaps with strong photocatalytic capabilities. These results demonstrate that the MoS 2 /MSe heterojunctions are expected to significantly improve the photocatalytic efficiency, suggesting the promising application prospect of MoS 2 /MSe heterojunctions in photocatalytic technology.
Two-dimensional van der Waals heterojunctions are seen as a powerful strategy to tune the electronic properties and enhance their performance in devices. In this study, we have systematically investigated the electronic properties and energy band alignments of MoS 2 /MSe (M = In, Ga) heterojunctions using first-principles calculations. The MoS 2 /InSe heterojunction exhibits a type-I band alignment, whereas the MoS 2 /GaSe heterojunction presents a type-II band characteristic. Furthermore, the MoS 2 /InSe heterojunction achieves a transition from type-I to type-II energy band alignment in the strain range of −3 % to −4 % biaxial strain. In contrast, the MoS 2 /GaSe heterojunction consistently maintains the type-II energy band alignment. In addition, the tensile strain effectively enhanced the optical absorption of MoS 2 /GaSe in the visible spectral interval, and a red-shift/blue-shift of the absorption peaks with increasing tensile/compressive strains was observed. The MoS 2 /InSe heterojunction at −3 % and the MoS 2 /GaSe heterojunction at −2 % to −1 % strain intervals exhibited suitable band gaps with strong photocatalytic capabilities. These results demonstrate that the MoS 2 /MSe heterojunctions are expected to significantly improve the photocatalytic efficiency, suggesting the promising application prospect of MoS 2 /MSe heterojunctions in photocatalytic technology.
期刊：
SPECTROCHIMICA ACTA PART A-MOLECULAR AND BIOMOLECULAR SPECTROSCOPY,2026年346:126928 ISSN：1386-1425
通讯作者：
Dai, Cong;Xu, Zhifeng
作者机构：
[Xu, Ying] College of Chemistry and Materials Science, Hunan Engineering Research Center for Monitoring and Treatment of Heavy Metals Pollution in the Upper Reaches of Xiangjiang River, Hengyang Normal University, Hengyang 421001, China;[Dai, Cong] College of Chemistry and Materials Science, Hunan Engineering Research Center for Monitoring and Treatment of Heavy Metals Pollution in the Upper Reaches of Xiangjiang River, Hengyang Normal University, Hengyang 421001, China. Electronic address: daicong@hynu.edu.cn;[Xu, Zhifeng] College of Chemistry and Materials Science, Hunan Engineering Research Center for Monitoring and Treatment of Heavy Metals Pollution in the Upper Reaches of Xiangjiang River, Hengyang Normal University, Hengyang 421001, China. Electronic address: 1760321030@qq.com
通讯机构：
[Dai, Cong; Xu, Zhifeng] C;College of Chemistry and Materials Science, Hunan Engineering Research Center for Monitoring and Treatment of Heavy Metals Pollution in the Upper Reaches of Xiangjiang River, Hengyang Normal University, Hengyang 421001, China. Electronic address:
关键词：
Dual-channel sensing;Hypochlorous acid;Organic cage;Ratiometric signal
摘要：
The high-performance detection of hypochlorous acid (HClO) is crucial for life and environmental health, but still challenging for the single fluorescence signal and single sensing channel of probes. Organic cage with strong adsorption ability and cage structure-related fluorescence property, is a type of potential fluorescent probe. However, the single signal and short emission of organic cage hinders its sensing performance. The design of dual emitting organic cage with long wavelength for dual-channel ratiometric sensing of HClO is of great significance. Herein, we reported an organic cage (ROX) as a probe for accurate quantification of HClO. The prepared ROX exhibited capsule topology-facilitated red/green dual emission around 660/502 nm. The cage-like structure was disrupted by the rapid specific redox interaction between imine bond in ROX and HClO. As a result, the fluorescence at 660 nm quenched in conjunction with the enhancement at 502 nm in PBS (pH 4). Besides, the fluorescence increased at 582 nm with constant emission at 480 nm in DMSO/PBS (pH 7) ( V / V , 9/1). The ratio signal in PBS and DMSO/PBS respectively gave the linearity range of 1–30 and 20–45 μM, the limit of detection (3 s ) of 14 and 49 nM, and the relative standard deviation of 1.04 % and 0.54 % for 10 replicate determinations. Based on the mutual validation of dual ratio, the dual-channel ratiometric sensing possesses good accuracy, sensitivity, selectivity, and precision. Moreover, ROX can be applied to HClO monitoring in actual water and disinfectant samples with the recoveries of spiked HClO ranged from 95.0 % to 104.2 %. The proposed sensing method highlights the great potential of organic cage in fluorescence sensing.
The high-performance detection of hypochlorous acid (HClO) is crucial for life and environmental health, but still challenging for the single fluorescence signal and single sensing channel of probes. Organic cage with strong adsorption ability and cage structure-related fluorescence property, is a type of potential fluorescent probe. However, the single signal and short emission of organic cage hinders its sensing performance. The design of dual emitting organic cage with long wavelength for dual-channel ratiometric sensing of HClO is of great significance. Herein, we reported an organic cage (ROX) as a probe for accurate quantification of HClO. The prepared ROX exhibited capsule topology-facilitated red/green dual emission around 660/502 nm. The cage-like structure was disrupted by the rapid specific redox interaction between imine bond in ROX and HClO. As a result, the fluorescence at 660 nm quenched in conjunction with the enhancement at 502 nm in PBS (pH 4). Besides, the fluorescence increased at 582 nm with constant emission at 480 nm in DMSO/PBS (pH 7) ( V / V , 9/1). The ratio signal in PBS and DMSO/PBS respectively gave the linearity range of 1–30 and 20–45 μM, the limit of detection (3 s ) of 14 and 49 nM, and the relative standard deviation of 1.04 % and 0.54 % for 10 replicate determinations. Based on the mutual validation of dual ratio, the dual-channel ratiometric sensing possesses good accuracy, sensitivity, selectivity, and precision. Moreover, ROX can be applied to HClO monitoring in actual water and disinfectant samples with the recoveries of spiked HClO ranged from 95.0 % to 104.2 %. The proposed sensing method highlights the great potential of organic cage in fluorescence sensing.
期刊：
Dyes and Pigments,2026年244:113133 ISSN：0143-7208
通讯作者：
Chuang Yao
作者机构：
[Yingde Niu; Fei Gu; Jinshan Wang] School of Materials Science and Engineering, Yancheng Institute of Technology, Yancheng, 224051, PR China;[Chuang Yao] Chongqing Key Laboratory of Extraordinary Bond Engineering and Advance Materials Technology (EBEAM), Yangtze Normal University, Chongqing, 408100, PR China;[Zhixin Dai; Yanhong Deng] College of Physics and Electronics Engineering, Hengyang Normal University, Hengyang, 421002, PR China;[Jianfeng Zhang] School of Integrated Circuits, Guangdong University of Technology, Guangzhou, 510006, PR China;[Yang Jiang] School of Materials Science and Engineering, Yancheng Institute of Technology, Yancheng, 224051, PR China<&wdkj&>Chongqing Key Laboratory of Extraordinary Bond Engineering and Advance Materials Technology (EBEAM), Yangtze Normal University, Chongqing, 408100, PR China
通讯机构：
[Chuang Yao] C;Chongqing Key Laboratory of Extraordinary Bond Engineering and Advance Materials Technology (EBEAM), Yangtze Normal University, Chongqing, 408100, PR China
摘要：
Through-space charge transfer (TSCT) thermally activated delayed fluorescence (TADF) emitters typically exhibit relatively low radiative decay rate ( k r s ), which remains a formidable obstacle to further improving device efficiency and suppressing efficiency roll-off. Here, two novel deep-blue TSCT-TADF emitters (YCIT14 and YCIT15) were rationally developed. Both emitters feature 2,4-diphenyl-1,3,5-triazine (DPTRz) and 2,4,6-triphenyl-1,3,5-triazine (TPTRz) as acceptor (A) units, which are interconnected via a phenyl bridging structure with 9-phenyl-9H-carbazole (9-PhCz) donor (D) group. Through precise structural modulation, YCIT15 incorporates an extra phenyl spacer unit compared to YCIT14, establishing a well-defined V-shaped conformation that enforces intimate face-to-face alignment between the donor and acceptor moieties. This distinct spatial configuration significantly amplifies intramolecular D-A electronic coupling through enhanced orbital overlap. These interactions enable not only through-bond charge transfer (TBCT) but also TSCT between the D and A fragments. As a consequence, YCIT15 achieves a small singlet-triplet energy gap (Δ E ST ) and high efficiency. Both molecules exhibit TSCT and TBCT characteristics, accompanied by rapid reverse intersystem crossing rates ( k RISC >10 5 s −1 ) and high radiative decay rate ( k r s >10 7 s −1 ) in neat films. The doped organic light-emitting diodes (OLEDs) utilizing YCIT14 and YCIT15 as emitters show deep-blue electroluminescence (EL), with emission peaks at 448 nm and 428 nm, respectively.
Through-space charge transfer (TSCT) thermally activated delayed fluorescence (TADF) emitters typically exhibit relatively low radiative decay rate ( k r s ), which remains a formidable obstacle to further improving device efficiency and suppressing efficiency roll-off. Here, two novel deep-blue TSCT-TADF emitters (YCIT14 and YCIT15) were rationally developed. Both emitters feature 2,4-diphenyl-1,3,5-triazine (DPTRz) and 2,4,6-triphenyl-1,3,5-triazine (TPTRz) as acceptor (A) units, which are interconnected via a phenyl bridging structure with 9-phenyl-9H-carbazole (9-PhCz) donor (D) group. Through precise structural modulation, YCIT15 incorporates an extra phenyl spacer unit compared to YCIT14, establishing a well-defined V-shaped conformation that enforces intimate face-to-face alignment between the donor and acceptor moieties. This distinct spatial configuration significantly amplifies intramolecular D-A electronic coupling through enhanced orbital overlap. These interactions enable not only through-bond charge transfer (TBCT) but also TSCT between the D and A fragments. As a consequence, YCIT15 achieves a small singlet-triplet energy gap (Δ E ST ) and high efficiency. Both molecules exhibit TSCT and TBCT characteristics, accompanied by rapid reverse intersystem crossing rates ( k RISC >10 5 s −1 ) and high radiative decay rate ( k r s >10 7 s −1 ) in neat films. The doped organic light-emitting diodes (OLEDs) utilizing YCIT14 and YCIT15 as emitters show deep-blue electroluminescence (EL), with emission peaks at 448 nm and 428 nm, respectively.
摘要：
The RAD7 detector is widely used for measuring the radon exhalation rate from the surfaces of media such as soil, rocks, and building materials. However, during the measurement process, the accuracy of the results is prone to interference due to the instrument's inherent statistical errors and environmental noise. To reduce these measurement errors, the Kalman filtering was introduced in this study to correct the radon exhalation rate, which was obtained through data fitting of radon concentration measured by the RAD7 detector. Ten verified experiments were performed with a radon exhalation standard device. The experimental result shows that 80 % of the radon exhalation rate, corrected by Kalman filtering, significantly approached the theoretical value of the standard device, compared to the uncorrected experimental results. It confirms the effectiveness of the Kalman filtering in correcting RAD7 measurements, thereby enhancing the accuracy of radon exhalation rate measurements. The proposed method provides a reference technical pathway for improving the measurement accuracy of similar radon measurement instruments.
The RAD7 detector is widely used for measuring the radon exhalation rate from the surfaces of media such as soil, rocks, and building materials. However, during the measurement process, the accuracy of the results is prone to interference due to the instrument's inherent statistical errors and environmental noise. To reduce these measurement errors, the Kalman filtering was introduced in this study to correct the radon exhalation rate, which was obtained through data fitting of radon concentration measured by the RAD7 detector. Ten verified experiments were performed with a radon exhalation standard device. The experimental result shows that 80 % of the radon exhalation rate, corrected by Kalman filtering, significantly approached the theoretical value of the standard device, compared to the uncorrected experimental results. It confirms the effectiveness of the Kalman filtering in correcting RAD7 measurements, thereby enhancing the accuracy of radon exhalation rate measurements. The proposed method provides a reference technical pathway for improving the measurement accuracy of similar radon measurement instruments.
作者：
Xiaoqing Wan*;Hui Liu;Feng Chen;Kun Hu;Zhize Li
期刊：
Digital Signal Processing,2026年168:105561 ISSN：1051-2004
通讯作者：
Xiaoqing Wan
作者机构：
[Hui Liu; Feng Chen; Kun Hu; Zhize Li] Hengyang Normal University, College of Computer Science and Technology, 421002, Hengyang, China;Hunan Provincial Key Laboratory of Intelligent Information Processing and Application, 421002, Hengyang, China;[Xiaoqing Wan] Hengyang Normal University, College of Computer Science and Technology, 421002, Hengyang, China<&wdkj&>Hunan Provincial Key Laboratory of Intelligent Information Processing and Application, 421002, Hengyang, China
通讯机构：
[Xiaoqing Wan] H;Hengyang Normal University, College of Computer Science and Technology, 421002, Hengyang, China<&wdkj&>Hunan Provincial Key Laboratory of Intelligent Information Processing and Application, 421002, Hengyang, China
摘要：
In recent years, the combination of convolutional neural networks (CNNs) with transformers for spectral-spatial feature extraction and robust semantic modeling has greatly improved the performance in hyperspectral image (HSI) classification tasks. However, these methods often overlook frequency information; CNNs struggle to capture global dependencies due to limited receptive fields, and transformers tend to lose fine-grained local structures and high-frequency variations. To address these challenges, this paper proposes a Laplacian frequency aware hierarchical network (LFAH-Net). We first design the method employing a diversity frequency-aware transformer (DFAT) module alongside a multi-level frequency fusion block (MFFB) stack to explicitly separate and integrate high-frequency signals such as edges and textures, as well as low-frequency signals like spectral contours, thereby achieving cross-level frequency feature complementarity. Besides, we propose a spectral-spatial adaptive recalibration fusion (SSARF) module, specifically designed to correct misalignments and suppress noise in hyperspectral features. Finally, the multi-scale dilation convolution (MSDC) module utilizes dilated convolutions to capture both local and global contextual information, while the adaptive feature fusion (AFF) module adaptively recalibrates and fuses these features with the spectral representations from DFAT. Experimental results on four popular hyperspectral datasets demonstrate that our framework significantly outperforms several state-of-the-art methods.
In recent years, the combination of convolutional neural networks (CNNs) with transformers for spectral-spatial feature extraction and robust semantic modeling has greatly improved the performance in hyperspectral image (HSI) classification tasks. However, these methods often overlook frequency information; CNNs struggle to capture global dependencies due to limited receptive fields, and transformers tend to lose fine-grained local structures and high-frequency variations. To address these challenges, this paper proposes a Laplacian frequency aware hierarchical network (LFAH-Net). We first design the method employing a diversity frequency-aware transformer (DFAT) module alongside a multi-level frequency fusion block (MFFB) stack to explicitly separate and integrate high-frequency signals such as edges and textures, as well as low-frequency signals like spectral contours, thereby achieving cross-level frequency feature complementarity. Besides, we propose a spectral-spatial adaptive recalibration fusion (SSARF) module, specifically designed to correct misalignments and suppress noise in hyperspectral features. Finally, the multi-scale dilation convolution (MSDC) module utilizes dilated convolutions to capture both local and global contextual information, while the adaptive feature fusion (AFF) module adaptively recalibrates and fuses these features with the spectral representations from DFAT. Experimental results on four popular hyperspectral datasets demonstrate that our framework significantly outperforms several state-of-the-art methods.
摘要：
Two-dimensional (2D) ferromagnetic materials integrating multiple functions are promising candidates for building magnetic and electronic nanodevices. Here, we predict a series of stable 2D multifunctional ferromagnetic monolayers VX (X = S, Se, Te) encompassing indirect semiconducting and half-metallic phases with sizable spin gaps. Due to the strong ferromagnetic coupling present in the VX monolayers, the magnetic transition temperatures ( Tc ) of VS, VSe, and VTe reach 369, 315, and 311 K, respectively. Furthermore, the magnetic and electronic properties of VX monolayers can be sensitively modulated via mechanical strain, while the VS and VSe monolayers further exhibit negative Poisson’s ratios. The VX monolayers thus represent an unusual family of 2D ferromagnetic materials with strong mechano-electromagnetic coupling that may serve as a building block for future multifunctional nanodevices.
摘要：
Currently, Camouflaged Object Detection (COD) methods often rely on single-view feature perception, which struggles to fully capture camouflaged objects due to environmental interference such as background clutter, lighting variations, and viewpoint changes. To address this, we propose the multi-view collaboration network (MCNet), inspired by human visual strategies for complex scene analysis. MCNet incorporates multiple perspectives for enhanced feature extraction. The global perception module takes the original, far, and near views, using different large-kernel convolutions and multi-head attention mechanisms for global feature embedding. In parallel, the local perception module processes the tilted, projected, and color-jittered views, extracting fine-grained local features through multi-branch deep convolutions and dilated convolutions. To facilitate deep interaction between global and local features, we introduce the hybrid interactive module, which explores the correlation of multi-view feature information and adaptively fuses features. For feature decoding, the dynamic pyramid shrinkage module integrates dynamic gated convolutions with a pyramid shrinkage mechanism, progressively aggregating semantic features through a hierarchical shrinking strategy and group fusion strategy. Experimental results on popular COD benchmark datasets show that MCNet outperforms 18 state-of-the-art methods.
Currently, Camouflaged Object Detection (COD) methods often rely on single-view feature perception, which struggles to fully capture camouflaged objects due to environmental interference such as background clutter, lighting variations, and viewpoint changes. To address this, we propose the multi-view collaboration network (MCNet), inspired by human visual strategies for complex scene analysis. MCNet incorporates multiple perspectives for enhanced feature extraction. The global perception module takes the original, far, and near views, using different large-kernel convolutions and multi-head attention mechanisms for global feature embedding. In parallel, the local perception module processes the tilted, projected, and color-jittered views, extracting fine-grained local features through multi-branch deep convolutions and dilated convolutions. To facilitate deep interaction between global and local features, we introduce the hybrid interactive module, which explores the correlation of multi-view feature information and adaptively fuses features. For feature decoding, the dynamic pyramid shrinkage module integrates dynamic gated convolutions with a pyramid shrinkage mechanism, progressively aggregating semantic features through a hierarchical shrinking strategy and group fusion strategy. Experimental results on popular COD benchmark datasets show that MCNet outperforms 18 state-of-the-art methods.
作者机构：
[Li, Lang] Hengyang Normal Univ, Coll Comp Sci & Technol, Hengyang 421002, Peoples R China.;Hengyang Normal Univ, Hunan Prov Key Lab Intelligent Informat Proc & App, Hengyang 421002, Peoples R China.
通讯机构：
[Li, L ] H;Hengyang Normal Univ, Coll Comp Sci & Technol, Hengyang 421002, Peoples R China.
关键词：
Side-channel analysis;Sample correlation locally;Deep learning;Kernel density estimation;Profiling analysis
摘要：
Label distribution learning techniques can significantly enhance the effectiveness of side-channel analysis. However, this method relies on using probability density functions to estimate the relationships between labels. The settings of parameters play a crucial role in the impact of the attacks. This study introduces a non-parametric statistical method to calculate the distribution between labels, specifically employing smoothing with the Gaussian kernel function and adjusting bandwidth. Then, the aggregation of the results from each label processed by the Gaussian kernel facilitates a hypothesis-free estimation of the label distribution. This method accurately represents the actual leakage distribution, speeding up guess entropy convergence. Secondly, we exploit similarities between profiling traces, proposing an analysis scheme for sample correlation locally of label distribution learning. Furthermore, Signal to-Noise Ratio (SNR) is employed to re-extract and reduce dataset dimensions to 500 power consumption points, resulting in noise reduction. Our results showcase the successful training of 800 profiling traces using our method for sample correlation locally of label distribution learning, with the findings indicating its exceptional performance.
Label distribution learning techniques can significantly enhance the effectiveness of side-channel analysis. However, this method relies on using probability density functions to estimate the relationships between labels. The settings of parameters play a crucial role in the impact of the attacks. This study introduces a non-parametric statistical method to calculate the distribution between labels, specifically employing smoothing with the Gaussian kernel function and adjusting bandwidth. Then, the aggregation of the results from each label processed by the Gaussian kernel facilitates a hypothesis-free estimation of the label distribution. This method accurately represents the actual leakage distribution, speeding up guess entropy convergence. Secondly, we exploit similarities between profiling traces, proposing an analysis scheme for sample correlation locally of label distribution learning. Furthermore, Signal to-Noise Ratio (SNR) is employed to re-extract and reduce dataset dimensions to 500 power consumption points, resulting in noise reduction. Our results showcase the successful training of 800 profiling traces using our method for sample correlation locally of label distribution learning, with the findings indicating its exceptional performance.
通讯机构：
[Zhao, HH ] H;Hengyang Normal Univ, Coll Comp Sci & Technol, Hengyang 421002, Hunan, Peoples R China.
关键词：
Text to image;Image generation;Generative Adversarial Network;Attention
摘要：
Text-to-image generation is a challenging and significant research task. It aims to synthesize high-quality images that match the given descriptive statements. Existing methods still have problems in generating semantic information fusion insufficiently, and the generated images cannot represent the descriptive statements properly. Therefore, A novel method named EMF-GAN (Efficient Multilayer Fusion Generative Adversarial Network) is proposed. It uses a Multilayer Fusion Module (MF Module) and Efficient Multi-Scale Attention Module (EMA Module) to fuse the semantic information into the feature maps gradually. It realizes the full utilization of the semantic information and obtains high-quality realistic images. Extensive experimental results show that our EMF-GAN is highly competitive in image generation quality and semantic consistency . Compared with the state-of-the-art methods, EMF-GAN shows significant performance improvement on both CUB (FID from 14.81 to 10.74) and COCO (FID from 19.32 to 16.86) datasets. It can generate photorealistic images with richer details and text-image consistency. Code can be found at https://github.com/zxcnmmmmm/EMF-GAN-master .
Text-to-image generation is a challenging and significant research task. It aims to synthesize high-quality images that match the given descriptive statements. Existing methods still have problems in generating semantic information fusion insufficiently, and the generated images cannot represent the descriptive statements properly. Therefore, A novel method named EMF-GAN (Efficient Multilayer Fusion Generative Adversarial Network) is proposed. It uses a Multilayer Fusion Module (MF Module) and Efficient Multi-Scale Attention Module (EMA Module) to fuse the semantic information into the feature maps gradually. It realizes the full utilization of the semantic information and obtains high-quality realistic images. Extensive experimental results show that our EMF-GAN is highly competitive in image generation quality and semantic consistency . Compared with the state-of-the-art methods, EMF-GAN shows significant performance improvement on both CUB (FID from 14.81 to 10.74) and COCO (FID from 19.32 to 16.86) datasets. It can generate photorealistic images with richer details and text-image consistency. Code can be found at https://github.com/zxcnmmmmm/EMF-GAN-master .
摘要：
Deep learning-based side-channel attacks (DL-SCA) are favored for their strong key recovery capabilities. However, their implementation is based on the attacker being able to manipulate a cloned device to build an attack model, which means that the attacker needs to know secret information in advance. The non-profiled side-channel attacks (NP-SCA) methods can complete the key recovery without knowing the secret information. Differential Deep Learning Analysis (DDLA) is the first NP-DLSCA method proposed in CHES2019, and several improved versions appeared later. In these methods, the bad quality of the raw traces, such as noise, random delay, etc., is often ignored, which limits the efficiency of key recovery. In this work, the conditional generative adversarial network (CGAN) is introduced and a novel framework NPSCA-CGAN is proposed to optimize traces in non-profiled SCA scenarios. We apply CGAN in non-profiled attacks and use plaintext to do trace labeling that optimizes the raw traces by training the generator to learn the label traces. The convolutional module and plaintext feature are added to the generator network to adapt various countermeasures. Moreover, a new traces quality evaluation metric average relative signal-to-noise ratio (AR-SNR) is proposed for non-profiled attack scenarios, which can directly reflect the performance of the traces in practical attack. The method is applied to unprotected, unaligned, and masked traces respectively. The experimental results indicate that it can enormously optimize the quality of the traces and improve the efficiency of non-profiled side-channel attacks.
作者机构：
[Ouyang, Wenxi; Liu, Sai; Fang, Yiling; Liu, Jinghua; Wang, Wenbin; Lu, Weihong; Lei, Ke; Li, Ying; Yuan, Chenxu; Yang, Guiying; Hong, Zuchen; Li, Zhongliang; Tan, Liang] Hengyang Normal Univ, Sch Chem & Mat Sci, Hunan Engn Res Ctr Monitoring & Treatment Heavy Me, Hengyang 421008, Peoples R China.;[Li, Hui] Nanjing Univ Sci & Technol, Inst Computat Mol & Mat Sci, Sch Chem & Chem Engn, Nanjing 210094, Peoples R China.;[Zheng, Qifeng; Cai, Yuepeng] South China Normal Univ, Sch Chem, Guangzhou Key Lab Mat Energy Convers & Storage, Guangzhou 510006, Peoples R China.
通讯机构：
[Cai, YP ] S;[Lu, WH ; Li, ZL] H;Hengyang Normal Univ, Sch Chem & Mat Sci, Hunan Engn Res Ctr Monitoring & Treatment Heavy Me, Hengyang 421008, Peoples R China.;South China Normal Univ, Sch Chem, Guangzhou Key Lab Mat Energy Convers & Storage, Guangzhou 510006, Peoples R China.
关键词：
aqueous zinc-ion batteries;dendrite-free;fluorine-decorated;homogenizing Zn2+ ion flux;metal-organic framework
摘要：
Aqueous zinc-ion batteries (AZIBs) have emerged as a promising complement to lithium-ion batteries due to their inherent safety benefits. However, the cycle life of AZIBs is severely limited by the poor stability of zinc anodes, manifested in uncontrolled dendritic growth and persistent side reactions, which hinder wider application. Herein, we report an ion-selective separator (UIO-66-4F/GF) achieved by in situ growth of a fluorine-functionalized metal–organic framework (UIO-66-4F) onto commercial glass fiber (GF). The synergistic mechanism, involving electrostatic repulsion between -F groups and SO 4 2 − $$ {\mathrm{SO}}_4^{2-} $$ anions along with strong interactions between -F and Zn 2+ cations, effectively restricts SO 4 2 − $$ {\mathrm{SO}}_4^{2-} $$ migration, suppresses 2D Zn 2+ diffusion across electrode interfaces, and enhances [Zn(H 2 O) 6 ] 2+ desolvation. Furthermore, the -F groups enable precise regulation of interfacial electric fields and Zn 2+ concentration gradients, thereby homogenizing ion flux to realize dendrite-free Zn deposition. The UIO-66-4F separator achieves stable Zn||Zn cell operation for 1500 h at 1 mA cm −2 via oriented deposition and sustains long-term cycling over 1000 h at 1 mA cm −2 , and delivers a Zn||Cu cell with 99.4% Coulombic efficiency. Moreover, the Zn|UIO-66-4F/GF|NH 4 V 4 O 10 full cell represents an ultrastable cycling stability with a high capacity retention of 90% after 500 cycles at a current density of 1 A g −1 .
