187037 research outputs found
Sort by
MONAQ: Multi-Objective Neural Architecture Querying for Time-Series Analysis on Resource-Constrained Devices
Get PDFThe growing use of smartphones and IoT devices necessitates efficient time-series analysis on resource-constrained hardware, which is critical for sensing applications such as human activity recognition and air quality prediction. Recent efforts in hardware-aware neural architecture search (NAS) automate architecture discovery for specific platforms; however, none focus on general time-series analysis with edge deployment. Leveraging the problem-solving and reasoning capabilities of large language models (LLM), we propose MONAQ, a novel framework that reformulates NAS into Multi-Objective Neural Architecture Querying tasks. MONAQ is equipped with multimodal query generation for processing multimodal time-series inputs and hardware constraints, alongside an LLM agent-based multi-objective search to achieve deployment-ready models via code generation. By integrating numerical data, time-series images, and textual descriptions, MONAQ improves an LLM's understanding of time-series data. Experiments on fifteen datasets demonstrate that MONAQ-discovered models outperform both handcrafted models and NAS baselines while being more efficient
I Was Told to Install the Antivirus App, but I'm Not Sure I Need It: Understanding Smartphone Antivirus Software Adoption and User Perceptions
No full textThe rising threat of mobile malware has prompted security vendors to recommend antivirus software for smartphones, yet user misconceptions, regulatory requirements, and improper use undermine its effectiveness. Our mixed-method study, consisting of in-depth interviews with 23 participants and a survey of 250 participants, examines smartphone antivirus software adoption in South Korea, where mandatory installation for banking and other financial apps is common. Many users confuse antivirus software with general security tools and remain unaware of its limited scope. Adoption is significantly influenced by perceived vulnerability, response efficacy, self-efficacy, social norms, and awareness, while concerns about system performance and skepticism about necessity lead to discontinuation or non-use. Mandatory installations for financial apps in South Korea contribute to user misconceptions, negative perceptions, and a false sense of security. These findings highlight the need for targeted user education, clearer communication about mobile-specific threats, and efforts to promote informed and effective engagement with antivirus software
CausalCFF: Causal Analysis between User Stress Level and Contextually Filtered Features Extracted from Mobile Sensor Data
No full textNowadays, it’s possible to deliver interventions through mobile technologies to improve users’ mental and physical health. Causal analysis may help researchers identify the potential causes of the health issues and design proper interventions. However, in previous studies, causal analysis is mainly conducted between single sensor data features such as walking activity duration and perceived stress. There is a lack of research into causal analysis between more complicated behavior features which could be derived from multiple sensor features and target well-being labels. To address this gap, we propose CausalCFF, a framework that investigates causal relationships between contextually filtered behavioral features (e.g., walking duration at workplace locations) and well-being outcomes (e.g., stress). Our analysis identifies frequent workplace visits during periods of reduced home time as the most salient cause for elevated stress levels, highlighting the framework’s ability to target context-specific behavioral biomarkers for human well-being. The code is also made availabl
Hera: A Heterogeneity-Aware Multi-Tenant Inference Server for Personalized Recommendations
No full textLanguage over Content: Tracing Cultural Understanding in Multilingual Large Language Models
No full textPolyphenolization: Hair Browning by Phenolic Adhesion, Ambient Oxygen, and Transition Metals
No full textSimilar to how astringency and spiciness persist on the tongue due to their strong anchoring to mucosal proteins, polyphenols demonstrate nonwashable properties. In this study, we investigated how polyphenol coating and subsequent oxidation, activated by ambient oxygen and transition metals, provide a convenient hair dyeing solution, such as apple browning. A synthetic polyphenol, dimeric 1,2,4-benzenetriol (diBTO), immediately coats to hair keratin, forming a nonwashable layer that resists detergent washing. The coated diBTO gradually oxidizes, exhibiting a brown color on the gray hair. Cu(II) ions serve as coordinating linkers between diBTO ligands, increasing coating stability and enhancing the brown color. Cysteine on the hair surface acts as an additional ligand, facilitating the rapid coating of polyphenols. This study reduces the long exposure time required for professional hair dyeing and demonstrates that gray hair care can be effectively achieved through polyphenol browning in daily life, offering a practical solution for maintaining hair color with regular shampooing.
STAGE: A compact and versatile TnpB-based genome editing toolkit for Streptomyces
No full textStreptomyces are naturally endowed with the capacity to produce a wide array of natural products with biomedical and biotechnological value. They have garnered great interest in synthetic biology applications given the abundance of uncharacterized biosynthetic gene clusters (BGCs). However, progress has been hindered by the limited availability of genetic tools for manipulating these bacteria. Several representative CRISPR-Cas systems have been established in Streptomyces to streamline experimental workflows and improve editing efficiency. Nevertheless, their broader applicability has been constrained by issues such as nuclease activity-related cytotoxicity and the large size of effector proteins. To address these challenges, we present Streptomyces-compatible TnpB-assisted genome editing (STAGE), a genetic toolkit based on ISDra2 TnpB, which is approximately one-third the size of Cas9 and enables precise, site-specific gene editing. We demonstrated that STAGE introduces genetic mutations with high efficiency and minimal off-target effects in two industrially important Streptomyces strains. Building on this platform, we developed STAGE-cBEST and STAGE-McBEST, enabling single and multiplexed C<middle dot>G-to-T<middle dot>A base editing, respectively, with editing efficiencies exceeding 75%. To further enhance performance, we engineered the ISDra2 TnpB system using an AI-assisted protein engineering framework, resulting in two variants that achieve nearly 100% genome editing efficiency. Additionally, through sequence homology analysis, we identified a TnpB ortholog from the same biological origin of ISDra2 TnpB, which also functions effectively as a gene editing tool. Our study establishes STAGE as a highly precise, programmable, and versatile genome editing platform for Streptomyces, paving the way for advanced genetic manipulation and synthetic biology applications in these industrially important bacteria.
Experimental validation of two-dimensional temperature tomography based on laser absorption spectroscopy in a flat flame burner
No full textThis study presents an experimental validation of a two-dimensional temperature reconstruction of a premixed methane/air flat flame using tunable diode laser absorption tomography (TDLAT). Two H2O absorption lines near 7185.6 and 7444.36 cm-1 were employed for two-line thermometry, with integrated absorbance data collected across a 5 x 5 laser beam grid. A modified algebraic reconstruction technique was used to retrieve spatially resolved temperature fields at various equivalence ratios. The reconstructed results revealed expected radial temperature distributions, with higher temperatures near the flame core and lower values toward the edges due to heat losses. Experimental validation was performed through comparison with coherent anti-Stokes Raman scattering measurements under identical conditions, demonstrating good agreement. Additional comparison with thermocouple data further emphasized the improved accuracy and non-intrusive nature of optical diagnostics. This work addresses a key limitation of previous TDLAT studies by providing direct validation through comparison with an independent optical technique. The results confirm the reliability of TDLAT for quantitative flame diagnostics and combustion research.
