30793 research outputs found
Sort by
Domain Tailored Large Language Models for Log Mask Prediction in Cellular Network Diagnostics
Get PDFSoftware logs generated by dedicated network testing hardware are often complex and bear minimal similarity to natural language, requiring the expertise of engineers to understand and capture defects recorded in these logs. This manual process is inefficient and expensive for both service providers and their clients. In this study, we demonstrate the transformative potential of Artificial Intelligence (AI), specifically through domain-tailoring of Large Language Models (LLMs) like RoBERTa, BigBird, and Flan-T5, to streamline the process of defect diagnostics. Particularly, we pre-train these models ground up on a real industrial telecommunications log corpus, and perform finetuning on a multi-label classification objective. This facilitates identifying a correct set of log points to be enabled for rapid detection of defects that arise during network testing. Despite encountering several challenges such as intricate text structures, heavily skewed label distribution, and inconsistencies in historical data labelling, our tailored LLMs achieve commendable performance on previously unseen defect cases, significantly reducing the turnaround times. This research not only serves as an exemplar for adapting LLMs in telecommunications industry for automated defect diagnostics, but also has wide implications for software log analysis across various industries
Tuning strength-ductility combination of the additively manufactured Al12Si based alloys via compositional regulation and plastic deformation
Get PDFData availability:
Data will be made available on request.Supplementary data are available online at: https://www.sciencedirect.com/science/article/pii/S0921509325000681?via%3Dihub#appsec1 .The enhanced strength-ductility synergy of Al12Si alloy manufactured via laser powder bed fusion (LPBF) was obtained through compositional regulation and post-fabrication plastic deformation. Microstructural characterizations showed that the typical synergy of multiple crystallised defects in brittle Si eutectic phase, including stacking faults (SFs), ultrafine nanotwins and 9R structure, co-contribute to the superb elongation of 23.7 % with the ultimate tensile strength of 189 MPa in as-LPBFed Al12Si alloy after hot-rolling deformation with a reduction rate of 70 %. In tandem with a trace of Mn and Fe additions, the UTS increases to 240 MPa with the El of 19.1 % under the same condition. Such an excellent strength-ductility synergy is mainly attributed to the formation of ultrafine α-Al12(Fe,Mn)3Si particles. This work offers a new perspective on the modification Al–Si alloys with the desired mechanical properties via LPBF and hot-rolling.Financial support from the National Natural Science Foundation of China (Grant No. 52071343) are gratefully acknowledged
A systematic literature review on incomplete multimodal learning: techniques and challenges
Get PDFData availability:
The data that support the findings of this study are available from the corresponding author, R.Y., upon reasonable request.Recently, machine learning technologies have been successfully applied across various fields. However, most existing machine learning models rely on unimodal data for information inference, which hinders their ability to generalize to complex application scenarios. This limitation has resulted in the development of multimodal learning, a field that integrates information from different modalities to enhance models' capabilities. However, data often suffers from missing or incomplete modalities in practical applications. This necessitates that models maintain robustness and effectively infer complete information in the presence of missing modalities. The emerging research direction of incomplete multimodal learning (IML) aims to facilitate effective learning from incomplete multimodal training sets, ensuring that models can dynamically and robustly address new instances with arbitrary missing modalities during the testing phase. This paper offers a comprehensive review of methods based on IML. It categorizes existing approaches based on their information sources into two main types: based on internal information and external information methods. These categories are further subdivided into data-based, feature-based, knowledge transfer-based, graph knowledge enhancement-based, and human-in-the-loop-based methods. The paper conducts comparative analyses from two perspectives: comparisons among similar methods and comparisons among different types of methods. Finally, it offers insights into the research trends in IML.This work is supported by National Natural Science Foundation of China (72401233), Jiangsu Provincial Qinglan Project, Natural Science Foundation of Jiangsu Higher Education Institutions of China (23KJB520038), Suzhou Science and Technology Programme (SYG202106), and Research Enhancement Fund of XJTLU (REF-23-01-008)
Right Move? Populist Radical Right Parties and Europe
Get PDFData access statement:
This article analyses existing data as cited in the “References” section.While before winning power populist radical right (PRR) parties are often overtly against the European Union (EU) and its policies, once in government their position can change significantly. Their threats to leave the EU are frequently replaced by requests for greater EU involvement, particularly in economic and migration issues. This Special Issue unpacks PRR parties’ discourse and policies in relation to Europe, before and after gaining power. Specifically, this Introduction outlines key theoretical debates on populism, the policy positioning of PRR parties on migration, foreign policy and other issues before and after entering government, and their relationship with the EU. It then addresses the performance of PRR parties in the 2024 European Parliament elections, and provides an overview of the Special Issue
Saccharin disrupts bacterial cell envelope stability and interferes with DNA replication dynamics
Get PDFData availability:
RNA-seq data: Gene Expression Omnibus GSE276752 and GSE238183. 3D-printed hydrogel mould blueprint: NIH 3D Printing Repository 3DPX-020380.
The source data of this paper are collected in the following database record: biostudies:S-SCDT-10_1038-S44321-025-00219-1.Saccharin has been part of the human diet for over 100 years, and there is a comprehensive body of evidence demonstrating that it can influence the gut microbiome, ultimately impacting human health. However, the precise mechanisms through which saccharin can impact bacteria have remained elusive. In this work, we demonstrate that saccharin inhibits cell division, leading to cell filamentation with altered DNA synthesis dynamics. We show that these effects on the cell are superseded by the formation of bulges emerging from the cell envelope, which ultimately trigger cell lysis. We demonstrate that saccharin can inhibit the growth of both Gram-negative and Gram-positive bacteria as well as disrupt key phenotypes linked to host colonisation, such as motility and biofilm formation. In addition, we test its potential to disrupt established biofilms (single-species as well as polymicrobial) and its capacity to re-sensitise multidrug-resistant pathogens to last-resort antibiotics. Finally, we present in vitro and ex vivo evidence of the versatility of saccharin as a potential antimicrobial by integrating it into an effective hydrogel wound dressing.RRMC and EM are supported by the NC3Rs PhD Studentship NC/V001582/1. RRMC and RD are supported by a Biotechnology and Biological Sciences Research Council New Investigator Award BB/V007823/1 and an MRC Award MR/Y001354/1. RRMC and CP are supported by the Academy of Medical Sciences/the Wellcome Trust/the Government Department of Business, Energy and Industrial Strategy/the British Heart Foundation/Diabetes UK Springboard Award (SBF006\1040). RD received a Research Mobility Award from the Young European Research University Network to collaborate with CSL and visit UiT. The work was supported by joint Research Grants BB/T007168/1 and BB/T006625-1 from the Biotechnology and Biological Sciences Research Council to CJR and ELB, as well as BBSRC research grant BB/W000393/1 to CJR. The work was further supported by a grant by the North Norwegian Health Trust (HelseNord, HN 1688-23) to CSL. MMH and ALW would like to thank the Tromsø Research Foundation and UiT Centre for New Antibacterial Strategies (CANS) for a start-up grant (TFS project ID: 18_CANS)
Culturing Human Pluripotent Stem Cells: Influence of Diverse Culture Conditions on Genomic Stability
No full text........
National Culture, Formal Institutions and Structure of Board of Directors: Theory and Empirical Evidence
Get PDFPurpose:
This study aims to explore how the structure of the board of directors is influenced by national informal culture values and the strength of formal institutional environments, as measured through legal regulations, market conditions and investor protection regulations.
Design/methodology/approach:
This study analyzes data from 432 companies listed in the S&P Global 1200 index using structural equation modeling. National cultural dimensions from Hofstede’s (2011) framework capture informal cultural aspects, while the World Bank’s Worldwide Governance Indicators assess formal institutions. This study examines board structure in terms of leadership style, board size, board independence, board committee structure and board diversity.
Findings:
The results reveal that national cultural values are negatively associated with rule of law institutions, indicating that culture can substitute for legal institutions, acting as “soft” regulation. Cultural values establish social norms and accountability when legal frameworks are weak. In addition, national culture positively relates to open market institutions, enhancing transparency, fairness and competition in strong markets. The findings also show that national culture and formal institutions significantly shape managerial perceptions of the board’s role and structure, impacting how firms prioritize monitoring versus resource provision.
Research limitations/implications:
The findings offer valuable insights for managers in diverse institutional contexts, enabling them to adjust board structures according to cultural and institutional factors.
Practical implications:
The research advocates for tailored governance practices that fit specific institutional and cultural contexts. Multinational corporations can benefit from customizing their governance structures according to the cultural and institutional environments of the countries in which they operate.
Originality/value:
This paper contributes to existing literature by focusing on complementarity as well as substitution mechanisms between national cultural characteristics and formal institutions in shaping board structure
Ultrasonographic evaluation of diaphragm fatigue in healthy humans
Get PDFData Availability Statement: The data that support the findings of this study are available from the corresponding author upon reasonable request.Assessment of diaphragm function and fatigue typically relies on the measurement of transdiaphragmatic pressure (Pdi). Although Pdi serves as an index of diaphragm force output, it provides limited information regarding the ability of the muscle to shorten and generate power. We asked whether ultrasonography, combined with Pdi, could be used to quantify changes in diaphragm function attributable to fatigue. Eight healthy men [mean (SD) age, 23 (7) years] completed two tasks on separate occasions: (i) 2 min of maximal isocapnic ventilation (MIV); or (ii) 3 × 5 min of maximal inspiratory resistive loading (IRL). Diaphragm function was evaluated before (PRE) and after each task (POST1, 10–15 min and POST2, 30–35 min) using synchronous recordings of Pdi and subcostal ultrasound traces of the right crural hemidiaphragm during anterolateral magnetic stimulation of the phrenic nerves and progressive CO2 rebreathing. Fatigue was quantified as pre- to post-loading changes in twitch Pdi, excursion velocity (excursion/time) and power (Pdi × velocity). Both tasks resulted in significant reductions in twitch Pdi (P < 0.05). There were no effects of MIV on ultrasound-derived measures. In contrast, IRL elicited a significant reduction in twitch excursion at POST1 (−16%; P = 0.034) and significant reductions in excursion velocity at POST1 (−32%; P = 0.022) and POST2 (−28%; P = 0.013). These reductions in excursion velocity, alongside the concurrent reductions in twitch Pdi, resulted in significant reductions in diaphragm power at POST1 (−48%; P = 0.009) and POST2 (−42%; P = 0.008). Neither task significantly altered the contractile responses to CO2. In conclusion, subcostal ultrasonography coupled with phrenic nerve stimulation is a promising method for quantifying contractile fatigue of the human diaphragm.None
Performance of subsurface drainage systems considering groundwater flow directions and network alignment
No full textData availability statement:
The original contributions presented in the study are included in the article/supplementary material; further inquiries can be directed to the corresponding authors.Groundwater is an important water supply for households, industry, and agriculture: one-third of the world’s drinking water is sourced from groundwater. The main objective of this study is to investigate the effects of groundwater flow directions on the design of drainage plans. The study involved field investigation to measure modeling input parameters and numerical modeling using the groundwater modeling system (GMS) code. It examined the effects of a subsurface drainage system, including net drainage recharge rate, equivalent drain depth, and drain spacing. This research was applied to the real-world case of the Mashtul Pilot Area (MPA) in Egypt. The results showed that the lateral arrangement had a high performance by orienting at 90°, 135°, and 45° angles. The groundwater levels reached 5.63, 5.33, and 5.51 m above mean sea level for the three orientations, respectively, at the baseline case. Also, the results showed that decreasing the drainage rates and increasing the equivalent drain depth have a greater effect on increasing groundwater levels, while increasing drain spacing led to an increase in the groundwater levels in the study. This approach is useful for the future arrangement of lateral subsurface drainage systems, which should be laid perpendicular to the direction of groundwater flow. In addition, the feasibility of this approach should be considered in the future design of this system. The results of the current study are useful for all stakeholders.The authors declare that no financial support was received for the research, authorship, and/or publication of this article
Ethnicity and health at work during the COVID-19
Get PDFData availability:
The data that has been used is confidential.JEL classification:
J080; I140; I110This paper explores how health-work-related illnesses and injuries have changed during the COVID-19 pandemic for different ethnic groups and by gender. We find that not all groups were affected in the same way. While almost all men in all ethnic groups were on average less likely to work during the pandemic period, women were more likely to work. We also find that Mixed Ethnic and Pakistani women who reported a higher probability of working in the reference week had a higher risk of illness/injuries at work. Meanwhile, White men and Other ethnic groups with a reduced probability of working during the pandemic had a lower risk of illness/injuries at work. Long-term illness varied by ethnicity and gender, with men experiencing a reduction and women an increase in physical and mental health issues. This research provides valuable insights into the multifaceted impact of the COVID-19 pandemic on the health and work patterns of different ethnic groups and gender. Understanding and identifying these disparities is crucial for formulating targeted policies aimed at mitigating adverse effects and promoting equitable outcomes in regional studies and urban economics.Nicodemo receives funding from Horizon Europe [grant number ES/T008415/1] and from the National Institute for Health Research Applied Research Collaboration Oxford and Thames Valley at Oxford Health NHS Foundation Trust
Data sources:
Access Repository Dashboard
Do you manage Open Research Online? Become a CORE Member to access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard! 👇