Browse Articles

Two-State, Nonvolatile Memristors and Light-Assisted Logic-In Operation from Diarylethene-Based Single-Molecule Junctions

A lightweight neural attention-based model for service chatbots

Abstract The growing demand for efficient service chatbots has led to the development of various deep learning techniques, such as generative neural attention-based mechanisms. However, existing attention processes often face challenges in generating contextually relevant responses. This study introduces a lightweight neural attention mechanism designed to enhance the scalability of service chatbots by integrating a scalar function into the existing attention score computation. While inspired by scaling practices in transformer models, the proposed scalar is tailored to seq2seq architectures to optimize the alignment sequences, resulting in improved context relevance and reduced resource requirements. To validate its effectiveness, the proposed model was evaluated on a real-world Customer Support Twitter dataset. Experimental results demonstrate a +0.82 BLEU-4 improvement and a 28% reduction in training time per epoch over the baseline. Moreover, the model achieves the target validation loss two epochs earlier, indicating faster convergence and improved training stability. Further experiments investigated activation functions and weight initializers integrated into the proposed model to identify optimal configurations that optimize the model’s performance. Comparative experimental results show that the proposed modifications significantly enhance response accuracy and contextual relevance. This lightweight attention mechanism addresses key limitations of existing attention mechanisms. Future work may extend this approach by combining it with transformer-based architectures to support broader sequence prediction tasks, including machine translation, recommender systems, and image captioning.

Phenotypic variation in otolith shape of American shad across eastern North American rivers

Abstract Otolith shape analysis has been widely applied to study population structure and environmental influences in various fish species. However, research on American shad (Alosa sapidissima) otolith morphology remains scarce, despite its potential to provide insights into population differentiation and environmental adaptation. This study analyses otolith contour shape from 1141 American shad collected between 2000 and 2023 across eleven large rivers from Canada to Florida. Using a wavelet transform framework based on the à trous algorithm and B3-spline wavelet, we quantified otolith shape variability and assessed its effectiveness for population discrimination. Principal Component Analysis revealed significant shape variation, with key differences in the rostrum, antirostrum, and posterior region. Wavelet analysis identified two primary otolith morphologies—upper and lower rostrum—geographically structured along a latitudinal gradient. A Multilayer Perceptron neural network successfully classified individuals with 90.9% accuracy, highlighting strong population differentiation, particularly in the St. Lawrence and Delaware rivers. Cluster analysis identified five morphotypes with distinct spatial distributions, suggesting a role for local environmental conditions in shaping otolith morphology. These findings underscore the utility of otolith shape analysis in deciphering population structure and highlight potential links between environmental variation and phenotypic plasticity in American shad.

Numerical solutions for norovirus epidemic spread: implications for public health control

Extended Spectrum Beta-Lactamase-Producing Bacterial Clones in West Africa: A Systematic Review and Meta-Analysis from a One Health Perspective

Enantioselective Construction of Aza-bicyclo[3.2.1]octane Skeletons via Substituted π-Allyl Palladium Dipoles Originated from Neutral η³-Propargyl Palladium Species

Cracking mechanism and acoustic emission characteristics of granite under varying thermal effects

Rethinking the Excited-State Redox Properties of Iron(III) Complexes for LMCT Photoredox Catalysis

Study on constitutive model of marine soft soil based on disturbed state concept

Cu Nanocluster Size Effect Inducing the Transformation of Polysulfides to Cu₂S/CuS for Durable Sodium Storage

Immune microenvironment and prognostic genes of triple-negative breast cancer in the context of transcriptome and single-cell sequencing

Phase Behavior of Light-Responsive Lyotropic Liquid Crystals for Molecular Solar Thermal Energy Storage

Amine-modified waste tires for CO2 capture and resource utilization

Acidic Additives Enable Tunable Low Temperature Side Chain Cleavage and Performance Enhancement in Organic Photovoltaics

Molecular-Scale Asymmetry Nanochannels for High-Efficiency Osmotic Energy Generation

Analysis of phytosterols encapsulated in pegylated liposomes

Abstract Dipalmitoylphosphatidylcholine (DPPC) lipids were encapsulated in PEGylated liposomes with free stigmasterol (ST), stigmasterol myristate (ME), and stigmasterol oleate (OE). Their quality was assessed using TEM, zeta potential, and hydrodynamic diameter measurements. The liposomes were heated to 60 °C and 180 °C. The degradation of stigmasterol and fatty acids was considered, as was derivative formation. The results show that the liposomes fulfilled their intended function. The ST liposomes were smallest, while the ME liposomes were similar in size to the OE liposomes. The degree of degradation of the compounds encapsulated in the liposomes depended on their structure. After heating the samples to 60 °C, the extent of stigmasterol degradation ranged from 3.5% in ST to 4.3% in ME and 6.5% in OE. After heating to 180 °C, the lowest level of stigmasterol degradation was observed for OE (7.3%), while degradation in ST and ME reached 13.4% and 10.1%, respectively. The high level of oxyphytosterols in all samples heated to 180 °C raised concerns. The oxyphytosterol (SOP) content of the liposomes heated to 60 °C ranged from 23.2 mg/g in those with free stigmasterol to 6.3 mg/g and 6.4 mg/g in the liposomes with stigmasterol myristate and stigmasterol oleate, respectively. After heating to 180 °C, the total SOP content was significantly higher, ranging from 88.7 mg/g for OE to 7.4 and 29.6 mg/g for ME and ST, respectively.

Nano-Bi@Hard Carbon Composite Anode for Sodium-Ion Batteries with 385 Wh L^–1 Volumetric Energy Density and Durability

Correction: 3D-printed porous titanium rods equipped with vancomycin-loaded hydrogels and polycaprolactone membranes for intelligent antibacterial drug release

Phenoxyimine Iron Aryl and Alkoxide Complexes as Reactive Intermediates in Iron-Catalyzed C(sp²)–C(sp³) Suzuki-Miyaura Cross-Coupling

Substrate stiffness regulates triple-negative breast cancer signaling through CXCR4 receptor dynamics

Abstract Biophysical properties of the extracellular matrix (ECM), such as mechanical stiffness, directly regulate behaviors of cancer cells linked to cancer initiation and progression. Cells sense and respond to ECM stiffness in the context of dynamic changes in biochemical inputs, such as growth factors and chemokines. While commonly studied as isolated inputs, mechanisms by which combined effects of mechanical stiffness and biochemical factors affect functions of cancer cells remain poorly defined. Using a combination of elastically supportive surface (ESS) culture dishes with defined stiffnesses and single-cell imaging, we report here that culturing cells on a stiff (28 kPa) versus soft (1.5 kPa) substrate increases CXCR4 and EGFR expression and promotes greater ligand-dependent internalization of CXCR4. In addition to increased CXCR4 expression, a stiff ECM also increases basal activation of Akt and ERK as well as signaling through these kinases in response to CXCL12-α and EGF and promotes migration of triple negative breast cancer (TNBC) cells. These data implicate receptor dynamics as a key mediator of Akt and ERK signaling as a mechanism for adverse effects of enhanced ECM stiffness on disease progression in TNBC.