期刊文献+
共找到4,967篇文章
< 1 2 249 >
每页显示 20 50 100
基于DeepSeek-R1 API的参数微调策略及本地化知识平台的设计与实现
1
作者 赵勇 《科技与创新》 2026年第1期37-41,45,共6页
探讨了通过优化参数和构建高效本地知识库来提升深度求索R1(DeepSeek-R1)性能的方法。研究验证了温度自适应策略及其他核心参数微调机制的有效性,显著提高了生成文本的质量与响应速度,并增强了输出内容的多样性和准确性。详细描述了如... 探讨了通过优化参数和构建高效本地知识库来提升深度求索R1(DeepSeek-R1)性能的方法。研究验证了温度自适应策略及其他核心参数微调机制的有效性,显著提高了生成文本的质量与响应速度,并增强了输出内容的多样性和准确性。详细描述了如何利用樱桃工作室(Cherry Studio)进行知识库搭建,包括基于Pro/北京人工智能研究院(Beijing Academy of Artificial Intelligence,BAAI)/BAAI通用嵌入模型3(BAAI General Embedding Model 3,BGE-M3)的内容管理、技术实现及安全性保障等方面。此外,还提出了针对不同应用场景的参数组合建议。未来工作将集中在负载管理优化、精细化参数调整、多模态支持等领域。 展开更多
关键词 deepseek-r1 参数优化 知识库构建 Cherry Studio
在线阅读 下载PDF
AI赋能呼吸临床教学培养新路径——以DeepSeek-R1大模型为例
2
作者 陶红艳 《甘肃科技》 2026年第1期51-56,共6页
文章以DeepSeek-R1大模型为技术基座,构建面向数字胜任力培养的呼吸系统整合临床教学模式的有效性与实施路径。通过构建“数据驱动-虚拟推演-循证闭环-人机协同”四位一体的教学体系,结合典型呼吸系统病例,开展以数字素养、系统思维与... 文章以DeepSeek-R1大模型为技术基座,构建面向数字胜任力培养的呼吸系统整合临床教学模式的有效性与实施路径。通过构建“数据驱动-虚拟推演-循证闭环-人机协同”四位一体的教学体系,结合典型呼吸系统病例,开展以数字素养、系统思维与临床决策为核心能力的教学实践。该模式显著提升学员在多系统整合诊断、循证决策及人机协作等方面的数字胜任力,教学满意度达94.7%。以AI大模型为依托的数字胜任力培养模式,有助于弥合传统医学教育与智慧医疗时代的能力鸿沟,具有较高的推广与应用价值。 展开更多
关键词 数字胜任力 人工智能 临床教学 呼吸系统疾病 deepseek-r1
在线阅读 下载PDF
基于DeepSeek-R1的生产调度优化新范式:以岛式装配为例
3
作者 朱家豪 黄思翰 +2 位作者 黄铭 王国新 阎艳 《计算机集成制造系统》 北大核心 2026年第2期413-424,共12页
以ChatGPT、DeepSeek为代表的大语言模型为智能制造提供了新动力。生产调度作为智能制造的重点研究方向,如何利用大语言模型提升调度方案获取效率是当前的热点。因此,提出一种基于DeepSeek-R1的生产调度优化新范式,从大模型工具化视角... 以ChatGPT、DeepSeek为代表的大语言模型为智能制造提供了新动力。生产调度作为智能制造的重点研究方向,如何利用大语言模型提升调度方案获取效率是当前的热点。因此,提出一种基于DeepSeek-R1的生产调度优化新范式,从大模型工具化视角探索生产调度问题求解的新思路。以新能源汽车岛式装配线的生产调度作为切入点,充分利用DeepSeek-R1的自然语言处理与代码生成能力引入,构建了包含提示词设计、建模求解和迭代优化3个阶段的调度优化流程。提示词设计阶段通过将岛式装配调度问题转化为结构化提示词输入,包括问题假设、约束、优化目标等;建模求解阶段,DeepSeek-R1根据提示词自动生成求解代码得到初始调度方案;迭代优化阶段,用户基于初始调度方案,通过校验-反馈循环得到符合问题假设与约束的最终方案。案例分析表明,该方法可以实现小时级别(约8小时)的岛式装配线生产调度优化求解效率,通过多算法对比,验证了DeepSeek-R1赋能调度优化的可行性与优势。 展开更多
关键词 智能制造 岛式装配 大语言模型 deepseek-r1 生产调度 智能优化
在线阅读 下载PDF
DNASE1L3 Mediates Hepatocellular Carcinoma Tumor Growth and Organoid Models via the Wnt/β-Catenin Signaling Pathway
4
作者 Shulong Zhang Yijun Zhao +5 位作者 Li Geng Feihong Song Li Feng Jun Jiang Qianqian Cai Fei Fan 《Oncology Research》 2026年第3期691-724,共34页
Background:Hepatocellular carcinoma(HCC)is a highly lethal malignancy driven by both intrinsic oncogenic pathways and immune microenvironmental regulation.Emerging evidence suggests that DNASE1L3 may influence tumor b... Background:Hepatocellular carcinoma(HCC)is a highly lethal malignancy driven by both intrinsic oncogenic pathways and immune microenvironmental regulation.Emerging evidence suggests that DNASE1L3 may influence tumor biology and immune responses;however,its specific roles in HCC progression and macrophage-mediated regulation remain unclear.This study aimed to elucidate the biological functions of DNASE1L3 in HCC and to determine how it modulates tumor behavior and immune interactions.Methods:Bioinformatics analyses of the GSE41804 and Cancer Genome Atlas-Liver Hepatocellular Carcinoma(TCGA-LIHC)datasets were used to identify hub genes.Functional assays assessed the impact of DNASE1L3 on HCC cell proliferation,migration,invasion,and cell cycle progression.The effects of DNASE1L3 on macrophage polarization and the Wnt/β-catenin signaling pathway were examined using a co-culture system.An HCC organoid model was established to further validate its regulatory function.Results:Eight prognostic signature genes were identified,with deoxyribonuclease I-like 3(DNase I-like 3)selected as the hub gene.DNASE1L3 overexpression suppressed HCC cell growth,inhibited migration and invasion,induced G1 arrest,and modulated epithelial-mesenchymal transition(EMT)markers.DNASE1L3 knockdown promoted M2-like macrophage polarization.Mechanistically,DNASE1L3 interacted withβ-catenin to enhance its ubiquitination and degradation,thereby inhibiting Wnt/β-catenin signaling and reducing PD-L1 expression.DNASE1L3 overexpression similarly restricted organoid growth and suppressed pathway activity.Conclusion:DNASE1L3 acts as a negative regulator of HCC progression by targeting the Wnt/β-catenin pathway and reducing PD-L1 expression,thereby influencing both tumor cell behavior and macrophage-mediated immune responses. 展开更多
关键词 Hepatocellular carcinoma DNASE1L3 Wnt/β-catenin signaling pathway organoid models tumor growth
暂未订购
DeepSeek-R1工业智能体技术剖析 被引量:2
5
作者 刘桐杰 刘奋民 +1 位作者 李戎 孙继超 《自动化博览》 2025年第3期54-57,共4页
本文主要对DeepSeek-R1工业智能体技术在工业智能中的应用进行了剖析。本文首先基于工业智能技术成熟度评价体系,对比了原生人工智能与工业应用要求差异,分析人工智能技术工业落地难的原因,然后分析了DeepSeek-R1工业应用技术的优势及... 本文主要对DeepSeek-R1工业智能体技术在工业智能中的应用进行了剖析。本文首先基于工业智能技术成熟度评价体系,对比了原生人工智能与工业应用要求差异,分析人工智能技术工业落地难的原因,然后分析了DeepSeek-R1工业应用技术的优势及局限性,接着提出了工业智能体的定义、等级划分及功能划分,最后讨论了DeepSeek-R1结合工业智能体技术助力工业智能升级的潜力。展望未来,DeepSeek-R1工业智能体有望提升工业智能水平,其在技术与应用方面前景广阔。 展开更多
关键词 deepseek-r1 工业智能体 工业智能等级 大模型技术 智能化升级
在线阅读 下载PDF
地下管网非开挖领域DeepSeek-R1融合多模态时序数据的故障预测模型优化研究
6
作者 徐猛 卓奕君 丁锐 《非开挖技术》 2025年第3期47-54,共8页
随着边缘计算在工业物联网中的广泛应用,地下管网故障诊断模型的轻量化部署成为亟待解决的关键问题。传统基于深度学习的诊断模型(如ResNet、Transformer)因参数量庞大,难以满足边缘设备严格的存储与实时性要求。本研究提出一种面向Deep... 随着边缘计算在工业物联网中的广泛应用,地下管网故障诊断模型的轻量化部署成为亟待解决的关键问题。传统基于深度学习的诊断模型(如ResNet、Transformer)因参数量庞大,难以满足边缘设备严格的存储与实时性要求。本研究提出一种面向DeepSeek-R1大语言模型的轻量化压缩框架,通过多粒度混合剪枝、硬件感知动态量化与编译时内存优化技术,实现模型参数量压缩至原尺寸的15%,推理速度提升3.2倍。实验表明,在合成管网数据集与UCR时序基准上,压缩后模型分类F1值达89.3%,推理延迟≤10ms,内存占用降低至312MB。本研究为边缘侧时序-空间联合特征建模提供了理论突破,并开源跨平台压缩工具链,支持ARM/RISC-V硬件自动适配。 展开更多
关键词 模型压缩 边缘计算 deepseek-r1 硬件感知量化 编译优化
在线阅读 下载PDF
基于DeepSeek-R1大语言模型的混凝土HJC模型优化方法
7
作者 宋帅 徐翔云 +2 位作者 任王军 房雨雨 姜金佐 《兵工学报》 北大核心 2025年第12期139-154,共16页
为探索基于DeepSeek-R1大语言模型优化混凝土本构模型的可行性及方法,采用DeepSeek-R1大语言模型对HJC模型理论进行了重构与改进,并提出了改进的HJC模型(DeepSeek-R1 Holmquist-Johnson-Cook,DSHJC_1)。研究系统总结了HJC模型的优缺点,... 为探索基于DeepSeek-R1大语言模型优化混凝土本构模型的可行性及方法,采用DeepSeek-R1大语言模型对HJC模型理论进行了重构与改进,并提出了改进的HJC模型(DeepSeek-R1 Holmquist-Johnson-Cook,DSHJC_1)。研究系统总结了HJC模型的优缺点,完成文本数据预处理与交互式优化流程,并提出相应本构模型的参数标定方法。通过单元测试和模拟爆炸作用下混凝土的损伤特性,对DSHJC-1和HJC模型的数值模拟结果进行比较分析,并通过模拟分离式霍普金森压杆(Split Hopkinson Pressure Bar,SHPB)试验,对DSHJC-1模型有效性展开进一步验证。研究结果表明:DeepSeek-R1大语言模型能够根据混凝土力学特性优化其本构关系的数学模型,但在物理意义的解释方面存在一定局限性,其推理的有效性依赖于高质量的文本与理论公式的混合数据输入;所提DSHJC-1模型改进了HJC模型在应变硬化、拉伸软化及拉伸裂纹描述等方面的不足,并可以有效模拟混凝土在爆炸冲击作用下的损伤特征。 展开更多
关键词 deepseek-r1大语言模型 deepseek-r1 Holmquist-Johnson-Cook模型 HJC模型 数值模拟 爆炸冲击
在线阅读 下载PDF
概念病症的人机对谈——和DeepSeek-R1互动的全记录
8
作者 夏传玲 《社会研究方法评论》 2025年第2期16-59,共44页
《概念病症的人机对谈——和DeepSeek-R1互动的全记录》是一篇非常有意思且严肃的文章。这篇文章记录了一次与当前广受关注的大语言模型的中方代表DeepSeek-R1(以下简称DS)对话的完整过程,讨论的主题是社会科学研究概念建构中存在的问... 《概念病症的人机对谈——和DeepSeek-R1互动的全记录》是一篇非常有意思且严肃的文章。这篇文章记录了一次与当前广受关注的大语言模型的中方代表DeepSeek-R1(以下简称DS)对话的完整过程,讨论的主题是社会科学研究概念建构中存在的问题。全文通过研究者与DS的六轮对话,展示了DS在解答复杂问题方面的惊人成就,为我们判断大语言模型当前的发展水平提供了极好的案例;同时本文也提供了一个使用精准的提示词高效引导DS更好地回答用户问题的范例。此外,文章还提出了三个值得读者深入思考的问题。第一,文章选取用来示范与DS讨论的主题,也即概念的建构问题,本身就是社会科学研究尤其是社会科学理论和方法中的根本性问题。在研究中建构概念看似平常,但实际上可以说是表征学术研究水平和学者功力的重要乃至终极标志,建构概念的过程体现了研究者对研究对象及其环境的全部认知,每一个概念背后都是一部本体论、认识论和方法论的历史。在这个对话过程中,作者展现了自身所具有的深厚理论和方法素养,而DS也显现出了极强的理解和应答能力;对话所传递的信息对社会科学研究者具有很好的启示价值,也有助于社会研究的初学者系统地思考概念建构问题。 展开更多
关键词 人机对谈 deepseek-r1 社会科学研究
在线阅读 下载PDF
基于3D-PDUS血流参数、血清sEng、sFlt-1构建孕中期孕妇胎盘功能不全风险的预测模型
9
作者 刘淑华 蔡洁 徐秋娟 《现代生物医学进展》 2026年第2期278-284,共7页
目的:探讨三维能量多普勒超声(three-dimensional power Doppler ultrasound,3D-PDUS)血流参数与血清可溶性内皮因子(soluble endoglins,sEng)、可溶性fms样酪氨酸激酶-1(soluble FMS-like tyrosine kinase-1,sFlt-1)联合检测对孕中期... 目的:探讨三维能量多普勒超声(three-dimensional power Doppler ultrasound,3D-PDUS)血流参数与血清可溶性内皮因子(soluble endoglins,sEng)、可溶性fms样酪氨酸激酶-1(soluble FMS-like tyrosine kinase-1,sFlt-1)联合检测对孕中期孕妇胎盘功能不全(placental insufficiency,PI)的预测价值,构建针对性风险预测模型。方法:本文采用前瞻性队列研究,选取2021年1月至2024年6月在江阴市人民医院进行规律产检的孕中期孕妇162例作为研究对象,依据孕晚期临床诊断分为PI组(74例)与非PI组(88例)。比较两组3D-PDUS血流参数及血清sEng、sFlt-1水平。通过单因素和多因素Logistic回归分析孕中期孕妇PI的影响因素;采用受试者工作特征(receiver operating characteristic,ROC)曲线分析3D-PDUS血流参数、血清sEng、sFlt-1单独及联合检测对孕中期孕妇PI发生风险的预测效能。结果:两组年龄、孕周、收缩压、舒张压、空腹血糖、胎儿双顶径、孕次、产次、既往流产史等一般资料比较,差异均无统计学意义(P>0.05),两组体质量指数(body mass index,BMI)比较,差异有统计学意义(P<0.05)。PI组血流指数(flow index,FI)、血管指数(vascular index,VI)及血管-血流指数(vascular-flow index,VFI)均显著低于非PI组,差异均有统计学意义(P<0.05)。PI组血清sEng、sFlt-1水平均显著高于非PI组,差异均有统计学意义(P<0.05)。多因素Logistic回归分析显示:sEng升高、sFlt-1升高是孕中期孕妇PI的独立危险因素(P<0.05),FI升高、VI升高、VFI升高是保护因素(P<0.05)。ROC曲线显示,3D-PDUS血流参数联合血清sEng、sFlt-1检测的曲线下面积(area under curve,AUC)为0.921,显著高于各指标单独检测(P<0.05)。结论:基于3D-PDUS血流参数与血清sEng、sFlt-1构建的联合模型对孕中期孕妇PI风险具有较高预测效能,可作为临床辅助筛查孕中期孕妇PI的工具。 展开更多
关键词 孕中期孕妇 胎盘功能不全 三维能量多普勒超声 血流参数 可溶性内皮因子 可溶性fms样酪氨酸激酶-1 预测模型
原文传递
MSCM:A geomagnetic model derived from Swarm,CSES,and MSS-1 satellite data and the evolution of the South Atlantic Anomaly 被引量:1
10
作者 Yu Gao ZhengTao Wang +2 位作者 Philip W.Livermore Hannah F.Rogers Cong Liu 《Earth and Planetary Physics》 2025年第3期564-576,共13页
Measurements from geomagnetic satellites continue to underpin advances in geomagnetic field models that describe Earth's internally generated magnetic field.Here,we present a new field model,MSCM,that integrates v... Measurements from geomagnetic satellites continue to underpin advances in geomagnetic field models that describe Earth's internally generated magnetic field.Here,we present a new field model,MSCM,that integrates vector and scalar data from the Swarm,China Seismo-Electromagnetic Satellite(CSES),and Macao Science Satellite-1(MSS-1)missions.The model spans from 2014.0 to 2024.5,incorporating the core,lithospheric,and magnetospheric fields,and it shows characteristics similar to other published models based on different data.For the first time,we demonstrate that it is possible to successfully construct a geomagnetic field model that incorporates CSES vector data,albeit one in which the radial and azimuthal CSES vector components are Huber downweighted.We further show that data from the MSS-1 can be integrated within an explicitly smoothed,fully time-dependent model description.Using the MSCM,we identify new behavior of the South Atlantic Anomaly,the broad region of low magnetic field intensity over the southern Atlantic.This prominent feature appears split into a western part and an eastern part,each with its own intensity minimum.Since 2015,the principal western minimum has undergone only modest intensity decreases of 290 nT and westward motion of 20 km per year,whereas the recently formed eastern minimum has shown a 2–3 times greater intensity drop of 730 nT with no apparent east-west motion. 展开更多
关键词 GEOMAGNETISM SWARM CSES MSS-1 geomagnetic field model
在线阅读 下载PDF
Elevated CXCL1 triggers dopaminergic neuronal loss in the substantia nigra of C57BL/6J mice:Evaluation of a novel Parkinsonian mouse model 被引量:1
11
作者 Xi-Zhen Ma Guo-Rui Jia +5 位作者 Meng-Yu Li Sheng-Han Zhang Zhao-Xin Wang Ning Song Ying-Juan Liu Jun-Xia Xie 《Zoological Research》 2025年第1期225-235,共11页
Substantial evidence points to the early onset of peripheral inflammation in the development of Parkinson's disease(PD),supporting the“body-first”hypothesis.However,there remains a notable absence of PD-specific... Substantial evidence points to the early onset of peripheral inflammation in the development of Parkinson's disease(PD),supporting the“body-first”hypothesis.However,there remains a notable absence of PD-specific animal models induced by inflammatory cytokines.This study introduces a novel mouse model of PD driven by the proinflammatory cytokine CXCL1,identified in our previous research.The involvement of CXCL1 in PD pathogenesis was validated using subacute and chronic MPTP-induced mouse models.Based on these findings,2-month-old C57BL/6J mice were intravenously administered CXCL1(20 ng/kg/day)for 2 weeks(5 days per week),successfully replicating motor deficits and pathological alterations in the substantia nigra observed in the chronic MPTP model.These results demonstrate the potential of CXCL1-induced inflammation as a mechanism for PD modeling.The model revealed activation of the PPAR signaling pathway in CXCL1-mediated neuronal damage by CXCL1.Linoleic acid,a PPAR-γactivator,significantly mitigated MPTPand CXCL1-induced toxicity and reduced serum CXCL1levels.In addition,the CXCL1-injected mouse model shortened the timeline for developing chronic PD mouse model to 2 weeks,offering an efficient platform for studying inflammation-driven processes in PD.The findings provide critical insights into the inflammatory mechanisms underlying PD and identify promising therapeutic targets for intervention. 展开更多
关键词 Parkinson’s disease Mouse model CXCL1 Inflammation PPAR signaling pathway
暂未订购
A 100-degree lithospheric magnetic field model constructed using MSS-1,Swarm-A,and CHAMP satellite data 被引量:1
12
作者 JiaXuan Zhang Yan Feng +3 位作者 Pan Zhang YuXuan Lin XinWu Li Ya Huang 《Earth and Planetary Physics》 2025年第3期667-676,共10页
By combining data from the Challenging Minisatellite Payload(CHAMP),Swarm-A,and newest Macao Science Satellite-1(MSS-1) missions,we constructed a lithospheric magnetic field model up to spherical harmonic degree N = 1... By combining data from the Challenging Minisatellite Payload(CHAMP),Swarm-A,and newest Macao Science Satellite-1(MSS-1) missions,we constructed a lithospheric magnetic field model up to spherical harmonic degree N = 100.To isolate the lithospheric magnetic field signals,we utilized the latest CHAOS-8(CHAMP,Φrsted,and SAC-C 8) model and MGFM(Multisource Geomagnetic Field Model) to remove nonlithospheric sources,including the core field,magnetospheric field,ocean tidal field,and ocean circulation field.Subsequently,orbit-by-orbit processing was applied to both scalar and vector data,such as spherical harmonic high-pass filtering,singular spectrum analysis,and line leveling,to suppress noise and residual signals along the satellite tracks.With an orbital inclination of only 41°,MSS-1 effectively captures fine-scale lithospheric magnetic field signals in mid-to low-latitude regions.Its data exhibit a root mean square error of only 0.77 nT relative to the final model,confirming the high quality and utility of lithospheric field modeling.The resulting model exhibits excellent consistency with the MF7(Magnetic Field Model 7),maintaining a high correlation up to N = 90 and still exceeding 0.65 at N = 100.These results demonstrate the reliability and value of MSS-1 data in global lithospheric magnetic field modeling. 展开更多
关键词 lithospheric magnetic field Macao Science Satellite-1(MSS-1) line leveling Magnetic Field model 7(MF7) singular spectrum analysis
在线阅读 下载PDF
人工智能赋能下的教学方法创新与实践研究——基于DeepSeek-R1建模的物理试题探讨
13
作者 林日兴 《中学物理教学参考》 2025年第27期4-8,共5页
针对传统物理实验教学中存在的资源不足、数据分析效率低下等问题,基于DeepSeek-R1人工智能技术,以“冷库照明电路改装”为例开展研究:通过建构照明灯L和指示灯D的Verilog-A电学模型,实现了非线性元件特性的精准模拟;采用“建模-仿真-... 针对传统物理实验教学中存在的资源不足、数据分析效率低下等问题,基于DeepSeek-R1人工智能技术,以“冷库照明电路改装”为例开展研究:通过建构照明灯L和指示灯D的Verilog-A电学模型,实现了非线性元件特性的精准模拟;采用“建模-仿真-验证”的方法,利用负载线分析和电阻阻值动态直流扫描验证了计算机CAD仿真工具,得出结果与理论计算值仅有0.02%的误差。指出该技术显著提升了实验分析效率,降低了学习难度,为教育资源均衡化提供了数字化解决方案,验证了人工智能在实验教学中的可行性与推广价值。 展开更多
关键词 deepseek-r1建模 人工智能 实验探究 物理教学
在线阅读 下载PDF
DeepSeek-R1模型在教学质量动态评估中的应用研究——基于教务管理系统的多维度数据分析
14
作者 赵军 《苏州工学院学报》 2025年第5期119-124,共6页
传统教学质量评估方式在数据处理深度与实时反馈能力方面存在明显不足,难以满足当前教育高质量发展的需求。本研究基于教务管理系统的实际运行环境,提出将DeepSeek-R1模型引入教学质量动态评估体系中,探索其在多维度数据融合与智能判别... 传统教学质量评估方式在数据处理深度与实时反馈能力方面存在明显不足,难以满足当前教育高质量发展的需求。本研究基于教务管理系统的实际运行环境,提出将DeepSeek-R1模型引入教学质量动态评估体系中,探索其在多维度数据融合与智能判别方面的应用潜力。通过对实践教学数据、多媒体教室使用记录、教师教学行为日志、学生反馈数据等进行深度嵌套分析,构建具有预测性与适应性的教学质量评估体系。研究表明,应用DeepSeek-R1模型能够有效提升教学质量监测的精准度与反馈效率,为教育管理部门实现精准施策提供强有力的智能支撑。 展开更多
关键词 教学质量评估 deepseek-r1 多维度数据 教务系统 人工智能
在线阅读 下载PDF
人工智能助力内部审计高质量发展的实践路径探析——以Deepseek-R1大语言模型应用为例
15
作者 彭雪妍 《国际商务财会》 2025年第23期86-89,共4页
生成式人工智能的兴起与发展,既深度重塑社会运行、生产模式与生活方式,也为审计行业迭代升级带来全新机遇与挑战。文章以国产人工智能——Deepseek-R1大语言模型为研究载体,深入剖析该模型在审计核心业务场景中的具体应用路径与实际效... 生成式人工智能的兴起与发展,既深度重塑社会运行、生产模式与生活方式,也为审计行业迭代升级带来全新机遇与挑战。文章以国产人工智能——Deepseek-R1大语言模型为研究载体,深入剖析该模型在审计核心业务场景中的具体应用路径与实际效用,以期为人工智能驱动内审工作高质量发展提供创新思路与多元视角,助力审计工作加速实现数字化、智能化转型升级。 展开更多
关键词 生成式人工智能 deepseek-r1应用 审计数字化
在线阅读 下载PDF
Comparison of the pathogenicity and neutrophil and monocyte response between SARS-CoV-2 prototype and Omicron BA.1 in a lethal mouse model
16
作者 Na Rong Jing Wu +6 位作者 Binbin Zhao Wanjun Peng Hekai Yang Gengxin Zhang Dangting Ruan Xiaohui Wei Jiangning Liu 《Animal Models and Experimental Medicine》 2025年第4期707-717,共11页
Background:SARS-CoV-2,first identified in late 2019,has given rise to numerous variants of concern(VOCs),posing a significant threat to human health.The emer-gence of Omicron BA.1.1 towards the end of 2021 led to a pa... Background:SARS-CoV-2,first identified in late 2019,has given rise to numerous variants of concern(VOCs),posing a significant threat to human health.The emer-gence of Omicron BA.1.1 towards the end of 2021 led to a pandemic in early 2022.At present,the lethal mouse model for the study of SARS-CoV-2 needs supplementation,and the alterations in neutrophils and monocytes caused by different strains remain to be elucidated.Methods:Human ACE2 transgenic mice were inoculated with the SARS-CoV-2 proto-type and Omicron BA.1,respectively.The pathogenicity of the two strains was evalu-ated by observing clinical symptoms,viral load and pathology.Complete blood count,immunohistochemistry and flow cytometry were performed to detect the alterations of neutrophils and monocytes caused by the two strains.Results:Our findings revealed that Omicron BA.1 exhibited significantly lower vir-ulence compared to the SARS-CoV-2 prototype in the mouse model.Additionally,we observed a significant increase in the proportion of neutrophils late in infection with the SARS-CoV-2 prototype and Omicron BA.1.We found that the proportion of monocytes increased at first and then decreased.The trends in the changes in the proportions of neutrophils and monocytes induced by the two strains were similar.Conclusion:Our study provides valuable insights into the utility of mouse models for simulating the severe disease of SARS-CoV-2 prototype infection and the milder manifestation associated with Omicron BA.1.SARS-CoV-2 prototype and Omicron BA.1 resulted in similar trends in the changes in neutrophils and monocytes. 展开更多
关键词 animal model SARS-CoV-2 Omicron BA.1 MONOCYTE NEUTROPHIL
暂未订购
Development of a capillary bundle evaporation advanced mathematical modeling for 1,2-propylene glycol-glycerin mixtures in porous media
17
作者 Bingbing Li Jiantong Li +3 位作者 Side Ren Shuo Gu Zhanjian Liu Liyan Liu 《Chinese Journal of Chemical Engineering》 2025年第4期261-273,共13页
Porous liquid-conducting micro-heat exchangers have garnered considerable attention for their role in efficient heat dissipation in small electronic devices.This demand highlights the need for advanced mathematical mo... Porous liquid-conducting micro-heat exchangers have garnered considerable attention for their role in efficient heat dissipation in small electronic devices.This demand highlights the need for advanced mathematical models to optimize the selection of mixed heat exchange media and equipment design.A capillary bundle evaporation model for porous liquid-conducting media was developed based on the conjugate mass transfer evaporation rate prediction model of a single capillary tube,supplemented by mercury injection experimental data.Theoretical and experimental comparisons were conducted using 1,2-propanediol-glycerol(PG-VG)mixtures at molar ratios of 1:9,3:7,5:5,and 7:3 at 120,150,and 180℃.The Jouyban-Acree model was implemented to enhance the evaporation rate predictions.For the 7:3 PG-VG mixture at 180℃under the experimental conditions of the thermal medium,the model's error reduced from 16.75%to 10.84%post-correction.Overall,the mean relative error decreased from 11.76%to 5.98%after correction. 展开更多
关键词 Evaporation in porous media Capillary bundle model 1 2-propylene glycol-glycerin Evaporation rate
在线阅读 下载PDF
The Dantzig Selector:Sparse Signals Recovery via l_(1-q)Minimization Model
18
作者 LI Jie DENG Chaohong LI Baode 《新疆大学学报(自然科学版中英文)》 2025年第1期14-23,共10页
We propose the Dantzig selector based on the l_(1-q)(1<q≤2)minimization model for the sparse signal recovery.First,we discuss some properties of l_(1-q)minimization model and give some useful inequalities.Then,we ... We propose the Dantzig selector based on the l_(1-q)(1<q≤2)minimization model for the sparse signal recovery.First,we discuss some properties of l_(1-q)minimization model and give some useful inequalities.Then,we give a sufficient condition based on the restricted isometry property for the stable recovery of signals.The l_(1-2)minimization model of Yin-Lou-He is extended to the l_(1-q)minimization model. 展开更多
关键词 Dantzig selector l_(1-q)minimization model signal recovery restricted isometry property
在线阅读 下载PDF
A novel carcinogenic mouse model by site-directed insertion of tandem human HRAS large DNA fragment into 15E1 site
19
作者 Susu Liu Yanwei Yang +10 位作者 Guitao Huo Hao Yang Zhao Chen Ling YuYa Wang Shijie Zhai Xiaowei Sun Wenda Gu Yuan Cao Wei Gong Sanlong Wang Changfa Fan 《Animal Models and Experimental Medicine》 2025年第11期1983-1996,共14页
Background:The precise insertion of large DNA fragments(>3–5 kb)remains one of the key obstacles in establishment of genetically modified murine models.Methods:A 21 kb large DNA fragment containing three tandemly ... Background:The precise insertion of large DNA fragments(>3–5 kb)remains one of the key obstacles in establishment of genetically modified murine models.Methods:A 21 kb large DNA fragment containing three tandemly linked copies of the human HRAS gene was inserted into the genome of C57BL/6J mouse,generating a mouse model designated as KI.C57-ras(or named NF-h HRAS).Whole-genome sequencing and Sanger sequencing were utilized to it confirm precise insertion and copy number.The stability of transgene expression among different generations was verified from multiple aspects using by digital PCR,western blot and DNA sequencing.To assess tumor susceptibility in the mouse model,N-Nitroso-N-methylurea(MNU)was administered at a dosage of 75 mg/kg.Histopathological examinations were conducted using hematoxylin and eosin(H&E)staining.Results:The HRAS DNA fragment was inserted into mouse chromosome 15E1 site,locating between 80623202 bp and 80625020 bp.NF-h HRAS mice exhibited stable inheritance and displayed consistent phenotypes across individuals.Moreover,this mouse model exhibited a high susceptibility to carcinogens.Upon administration of MNU the earliest mortality onset was earlier than that of wild-type littermates(day 65 vs.day 78 for male and day 56 vs.day 84 for female).Notably,100%of the NF-h HRAS transgenic mice developed tumors,with approximately 84%of male NF-h HRAS mice exhibiting specific tumor types,such as squamous cell carcinoma or squamous cell papilloma,which was consistent with the previously reported carcinogenic rasH2 mouse model.The types of tumors and the target organs exhibited diversity in NFh HRAS mice,while the spontaneous tumor incidence remained low(1/50).Conclusions:The NF-h HRAS mice demonstrated excellent genetic stability,a reproducible phenotype,and high susceptibility to carcinogens,indicating their potential utility in non-clinical safety evaluations of drugs as per the S1B guidelines issued by the ICH(The International Council for Harmonization of Technical Requirements for Pharmaceuticals for Human Use). 展开更多
关键词 human HRAS gene ICH S1B guideline KI.C57-ras carcinogenic mouse model large DNA fragment editing non-clinical carcinogenicity evaluation
暂未订购
武威农村地区“1+2+N”模式对学前儿童语言表达能力提升效果的调研
20
作者 樊超 段晓静 《黑龙江科学》 2026年第1期77-79,共3页
为破解农村地区学前儿童语言表达能力发展滞后难题,验证“1+2+N”模式对学前儿童语言表达能力的提升效果,选取武威市3个县区6所农村幼儿园的240名3~6岁儿童为研究对象,综合运用实验法、问卷调查法及访谈法开展为期7个月的干预实验并进... 为破解农村地区学前儿童语言表达能力发展滞后难题,验证“1+2+N”模式对学前儿童语言表达能力的提升效果,选取武威市3个县区6所农村幼儿园的240名3~6岁儿童为研究对象,综合运用实验法、问卷调查法及访谈法开展为期7个月的干预实验并进行前后测对比研究。结果显示,实施“1+2+N”模式后,实验组儿童的语言表达能力总分提升21.43分,显著高于对照组,在词汇表达、语法运用、语用沟通三大核心维度均取得跨越式进步。其中,4~5岁儿童的提升率最高。N类课程中,家园共育类课程与儿童语言能力的相关性最强。“1+2+N”模式可有效弥补农村学前教育资源短板,对提升儿童语言表达能力具有显著效果,为民族地区及农村地区学前儿童语言发展提供了可复制、可推广的实践方案。 展开更多
关键词 武威农村 1+2+N”模式 学前儿童 语言表达能力 农村学前教育
在线阅读 下载PDF
上一页 1 2 249 下一页 到第
使用帮助 返回顶部