[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-5104":3,"related-5104":101},{"id":4,"title":5,"url":6,"summary":7,"summary_zh":8,"content":9,"source_name":10,"source_url":6,"published_at":11,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":15,"score_detail":16,"sources":23,"tags":25,"primary_topic":31,"secondary_topics":37,"classification":38,"search_phrases":73,"slug":76,"view_count":77,"doi":78,"paper":79,"created_at":100},5104,"Predicting Wetland Nitrogen Levels Through Microbial Signatures: A Taxonomy-Based Approach Validated by Machine Learning","https:\u002F\u002Fdoi.org\u002F10.14288\u002Fujemiplus.v11.203845","Wetland ecosystems harbour diverse microbial communities that play a key role in regulating biogeochemical cycles, yet their potential to predict abiotic soil conditions remains underexplored. This study investigates whether microbial community composition can be used to infer environmental conditions, focusing on five abiotic factors: pH, calcium, total carbon, respiration, and total nitrogen. Using 16S rRNA amplicon sequencing data processed in QIIME2, total nitrogen emerged as the only abiotic factor significantly associated with microbial alpha and beta diversity. Low-nitrogen wetlands exhibited greater microbial richness and evenness, while high-nitrogen soils were dominated by fewer, specialized taxa. Core microbiome and indicator species analyses identified eight indicator species that were uniquely linked to either high- or low-nitrogen conditions, reflecting distinct microbial signatures tied to nitrogen availability. These indicator species were used to train a Random Forest model that classified nitrogen levels with high levels of accuracy. Overall, microbial signatures proved to be reliable bioindicators of nitrogen conditions, and this machine learning-driven, taxonomy-based framework offers practical applications in ecological monitoring, precision agriculture, and sustainable land management.","湿地生态系统孕育着多样化的微生物群落，这些群落在调控生物地球化学循环中发挥着关键作用，然而其预测非生物土壤条件的潜力仍未得到充分探索。本研究探讨微生物群落组成能否用于推断环境条件，聚焦于五个非生物因子：pH、钙、总碳、呼吸作用和总氮。利用在QIIME2中处理的16S rRNA扩增子测序数据，总氮成为唯一与微生物α多样性和β多样性显著相关的非生物因子。低氮湿地表现出更高的微生物丰富度和均匀度，而高氮土壤则由少数特化类群主导。核心微生物组和指示物种分析鉴定出八个指示物种，它们分别与高氮或低氮条件独特关联，反映了与氮有效性相关的不同微生物特征。这些指示物种被用于训练随机森林模型，该模型以高准确率对氮水平进行分类。总体而言，微生物特征被证明是氮条件的可靠生物指示物，这种机器学习驱动的、基于分类学的框架为生态监测、精准农业和可持续土地管理提供了实际应用。",null,"Open Collections","2026-10-15T00:00:00Z","论文",10,false,73,{"impact":17,"substance":18,"depth":19,"authority":17,"freshness":20,"relevant":21,"comment":22},12,22,18,9,1,"方法新颖、结论可靠，将微生物组与机器学习结合用于湿地氮素监测，对农业面源污染与精准农业有参考价值，但属基础研究、影响范围有限。",[24],{"name":10,"url":6},[26,27,28,29,30],"农业人工智能","微生物组","氮素监测","生态监测","湿地生态",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},"I2","agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai","I",[],{"status":39,"locked":14,"version":40,"method":41,"evidence":42},"classified","2026-10-22.2","rules",{"candidates":43,"agricultural_evidence":46},[44,52,60,67],{"code":32,"score":45,"supported":46,"title":47,"abstract":49,"tags":51},15,true,[48],"machine learning",[50,48],"机器学习",[26],{"code":53,"score":54,"supported":46,"title":55,"abstract":56,"tags":59},"H2",6,[],[57,58],"土壤","soil",[],{"code":61,"score":54,"supported":46,"title":62,"abstract":63,"tags":66},"I1",[],[64,65],"精准农业","precision agriculture",[],{"code":68,"score":69,"supported":14,"title":70,"abstract":71,"tags":72},"H6",2,[],[],[30,29],[74,75],"16S rRNA 微生物 氮素","随机森林 湿地","16SrRNA微生物氮素-5104",0,"10.14288\u002Fujemiplus.v11.203845",{"doi":78,"openalex_id":80,"authors":81,"venue":10,"cited_by_count":77,"oa_url":6,"card":92,"direction":98,"ingested_from":99},"W7222166649",[82,84,86,88,90],{"name":83,"orcid":9},"Nadia Arjomandi",{"name":85,"orcid":9},"Yasmin Faghir",{"name":87,"orcid":9},"Ingrid Wang",{"name":89,"orcid":9},"Michelle Benitez-Muller",{"name":91,"orcid":9},"Ava Tiefenbach",{"tldr":93,"method":94,"finding":95,"direction":96,"opportunity":97},"利用湿地微生物群落特征结合机器学习预测土壤总氮水平。","16S rRNA测序、QIIME2分析、核心微生物组与随机森林模型。","总氮是唯一与微生物多样性显著相关的因子，8个指示种可高精度分类氮水平。","农业绿色发展与碳","可将微生物指示种与机器学习结合，开发低成本湿地氮素监测工具并推广至农田。","农业人工智能与决策模型","openalex","2026-10-16T23:30:46.851833Z",{"total":54,"page":21,"page_size":54,"items":102},[103,171,249,312,374,426],{"id":104,"title":105,"url":106,"summary":107,"summary_zh":108,"content":9,"source_name":109,"source_url":106,"published_at":110,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":111,"score_detail":112,"sources":115,"tags":117,"primary_topic":121,"secondary_topics":126,"classification":128,"search_phrases":152,"slug":155,"view_count":113,"doi":156,"paper":157,"created_at":170},3618,"Multi-Target Remote Sensing Segmentation with Cross-Scale Attention for Urban-Ecological Intelligence","https:\u002F\u002Fdoi.org\u002F10.56201\u002Fijgem.vol.12.no1.2026.pg131.154","Remote sensing image segmentation is essential to extract valuable information from satellite and aerial images to achieve significant applications such as urban planning and ecological monitoring. Yet, it is hard to accurately segment diverse and complicated features because of the constraints of conventional approaches and the requirement to process variable object sizes and complicated boundaries. This research aims to address the above difficulties through the use and assessment of advanced deep learning models specifically UNet, SegNet, and TransU-Net in multi target semantic segmentation. These networks have been applied in single-object segmentation tasks with a focus on structures and roads, while the UNet architecture was additionally utilized for multi-object segmentation comprising buildings, woods, grasslands, water bodies, and cultivated land. Using appropriate remote sensing datasets, the accuracy of the models was thoroughly evaluated using common evaluation metrics, including pixel accuracy, Intersection over Union (IoU), and F1-score. The experimental outcomes illustrate the capabilities of these deep learning methods to provide accurate identification of essential features. Notably, the maximum attainable accuracy for single-object building segmentation was 96.85%, whereas the overall accuracy for multi-object segmentation with the UNet was 84.2%. The experimental results show these deep learning algorithms can effectively separate different targets from remote sensing imagery, thus making them fundamental tools for geospatial analysis and related fields.","遥感图像分割对于从卫星和航空图像中提取有价值信息，以实现城市规划和生态监测等重要应用至关重要。然而，由于传统方法的局限以及处理可变目标尺寸和复杂边界的需要，准确分割多样且复杂的地物十分困难。本研究旨在通过使用和评估先进的深度学习模型，特别是UNet、SegNet和TransU-Net在多目标语义分割中的应用，来解决上述难题。这些网络已被应用于单目标分割任务，重点关注建筑物和道路，而UNet架构还被额外用于多目标分割，包括建筑物、林地、草地、水体和耕地。利用适当的遥感数据集，通过常用评价指标对模型的精度进行了全面评估，包括像素精度、交并比（IoU）和F1分数。实验结果表明了这些深度学习方法在准确识别关键地物方面的能力。值得注意的是，单目标建筑物分割的最高可达精度为96.85%，而使用UNet进行多目标分割的总体精度为84.2%。实验结果表明，这些深度学习算法能够有效地从遥感图像中分离出不同目标，从而使其成为地理空间分析及相关领域的基础工具。","IIARD INTERNATIONAL JOURNAL OF GEOGRAPHY AND ENVIRONMENTAL MANAGEMENT","2026-09-25T00:00:00Z",63,{"impact":17,"substance":19,"depth":45,"authority":13,"freshness":113,"relevant":21,"comment":114},8,"将UNet、SegNet、TransU-Net用于多目标遥感语义分割，建筑单目标精度96.85%、多目标84.2%，方法常规但结论对农业与生态遥感监测有参考价值。",[116],{"name":109,"url":106},[26,118,119,120,29],"深度学习","语义分割","遥感",{"code":122,"slug":123,"name":124,"path":125,"parent_code":36},"I3","remote-sensing","遥感与农情监测","\u002Ftopic\u002Fdigital-agriculture\u002Fremote-sensing",[127],{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},{"status":39,"locked":14,"version":40,"method":41,"evidence":129},{"candidates":130,"agricultural_evidence":46},[131,136,141,148],{"code":122,"score":45,"supported":46,"title":132,"abstract":134,"tags":135},[133],"remote sensing",[120,133],[120],{"code":32,"score":20,"supported":46,"title":137,"abstract":138,"tags":140},[],[118,139],"deep learning",[26,118,119],{"code":142,"score":143,"supported":46,"title":144,"abstract":145,"tags":147},"I4",3,[],[146],"空间分析",[],{"code":68,"score":21,"supported":14,"title":149,"abstract":150,"tags":151},[],[],[29],[153,154],"UNet 遥感 多目标分割","TransU-Net 城市生态 遥感","UNet遥感多目标分割-3618","10.56201\u002Fijgem.vol.12.no1.2026.pg131.154",{"doi":156,"openalex_id":158,"authors":159,"venue":109,"cited_by_count":77,"oa_url":162,"card":163,"direction":169,"ingested_from":99},"W7214298981",[160],{"name":161,"orcid":9},"Chibueze Favour Aririguzo","https:\u002F\u002Fiiardjournals.org\u002Fget\u002FIJGEM\u002FVOL. 12 NO. 1 2026\u002FMulti-Target Remote Sensing Segmentation 131-154.pdf",{"tldr":164,"method":165,"finding":166,"direction":167,"opportunity":168},"评估UNet、SegNet和TransU-Net在多目标遥感语义分割中的性能，用于城市与生态监测。","使用UNet、SegNet、TransU-Net模型，在遥感数据集上以像素精度、","单目标建筑分割最高精度96.85%，UNet多目标分割总体精度84.2%。","农业遥感与作物表型","可探索跨尺度注意力机制提升多目标分割精度，并迁移至耕地与作物精细分类。","智慧农业 \u002F 农业物联网","2026-09-27T23:30:28.366758Z",{"id":172,"title":173,"url":174,"summary":175,"summary_zh":9,"content":9,"source_name":176,"source_url":174,"published_at":177,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":178,"score_detail":179,"sources":185,"tags":187,"primary_topic":190,"secondary_topics":195,"classification":197,"search_phrases":218,"slug":221,"view_count":143,"doi":222,"paper":223,"created_at":248},2287,"UAV-based rice nitrogen status monitoring using machine learning and deep learning","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs11119-026-10440-8","UAV-based rice nitrogen status monitoring using machine learning and deep learning。Precision Agriculture","Precision Agriculture","2026-09-11T00:00:00Z",75,{"impact":180,"substance":181,"depth":182,"authority":183,"freshness":113,"relevant":21,"comment":184},16,20,17,14,"核心期刊论文，将机器学习与深度学习用于无人机水稻氮素诊断，方法新颖、结论可靠，对精准施肥有参考价值，但属细分技术进展，未达重大突破层级。",[186],{"name":176,"url":174},[188,26,189,120,28],"智慧农业","水稻",{"code":191,"slug":192,"name":189,"path":193,"parent_code":194},"A1","rice","\u002Ftopic\u002Ffield-crops\u002Frice","A",[196],{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},{"status":39,"locked":14,"version":40,"method":41,"evidence":198},{"candidates":199,"agricultural_evidence":46},[200,205,209,214],{"code":32,"score":201,"supported":46,"title":202,"abstract":203,"tags":204},23,[48,139],[48,139],[26],{"code":191,"score":17,"supported":46,"title":206,"abstract":207,"tags":208},[192],[192],[189],{"code":61,"score":210,"supported":46,"title":211,"abstract":212,"tags":213},4,[],[65],[188],{"code":122,"score":21,"supported":14,"title":215,"abstract":216,"tags":217},[],[],[120],[219,220],"农业人工智能 智慧农业 氮素监测 水稻","农业人工智能 智慧农业","农业人工智能智慧农业氮素监测水稻-2287","10.1007\u002Fs11119-026-10440-8",{"doi":222,"openalex_id":224,"authors":225,"venue":176,"cited_by_count":77,"oa_url":9,"card":9,"direction":167,"ingested_from":99},"W7212268243",[226,228,231,233,235,238,240,242,245],{"name":227,"orcid":9},"Xi Tao",{"name":229,"orcid":230},"Jiayi Zhang","https:\u002F\u002Forcid.org\u002F0000-0002-0401-996X",{"name":232,"orcid":9},"Zhaopeng Fu",{"name":234,"orcid":9},"Xinye Xu",{"name":236,"orcid":237},"Qiang Cao","https:\u002F\u002Forcid.org\u002F0000-0003-3733-2968",{"name":239,"orcid":9},"Yongchao Tian",{"name":241,"orcid":9},"Yan Zhu",{"name":243,"orcid":244},"Weixing Cao","https:\u002F\u002Forcid.org\u002F0000-0003-2622-7986",{"name":246,"orcid":247},"Xiaojun Liu","https:\u002F\u002Forcid.org\u002F0000-0001-7593-085X","2026-09-13T23:30:03.209118Z",{"id":250,"title":251,"url":252,"summary":253,"summary_zh":254,"content":9,"source_name":255,"source_url":252,"published_at":256,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":257,"score_detail":258,"sources":261,"tags":263,"primary_topic":267,"secondary_topics":273,"classification":275,"search_phrases":292,"slug":295,"view_count":77,"doi":296,"paper":297,"created_at":311},5592,"Evaluation of Concept-Rank Collapse Audits for Computer Vision Agriculture Disease Classifiers Under Low-Label Regimes: An Interpretability Capacity Bound for Smallholder Fungicide Selectivity","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23545193","Concept-based interpretability methods for plant-disease classifiers are typically evaluated by whether their concept activations correlate with human-annotated labels, but this criterion is silent on a prior question: whether the classifier's internal representation retains enough concept-rank structure for any audit to be informative. the present analysis formalises that prior question as an interpretability capacity bound. A classifier is modelled as a concept-rank operator whose singular values govern how much concept-level signal survives the projection from pixel space to class logits, and a collapse audit is defined as the ratio of retained concept-rank energy to total concept energy. Under a low-label regime, where concept annotations are scarce and the audit must be estimated from few labelled examples, the bound predicts that audit reliability degrades as a function of the concept-rank spectrum and the label budget. The analysis is a theoretical and methodological contribution: it derives the bound, states the conditions under which a collapse audit is identifiable, and characterises the operating characteristics of the audit as a decision procedure under a declared variance structure (between-cluster SD = 6.0, residual SD = 4.5, intraclass correlation = 0.64). Power was estimated across 2000 design cells for cluster counts of 8, 16, 24 and 32 and effect sizes of 3, 5 and 8 percentage points; 52 clusters were required for 5 percentage points at 80% power. The bound reframes concept audits as capacity tests rather than fidelity tests, with consequences for how smallholder fungicide-selectivity decisions should be supported by interpretable disease classifiers.","面向植物病害分类器的基于概念的可解释性方法，通常以其概念激活是否与人工标注标签相关作为评价标准，但该标准对一个更先前的问题保持沉默：分类器的内部表征是否保留了足够的概念秩结构，从而使任何审计具有信息量。本分析将该先前问题形式化为一个可解释性容量上界。分类器被建模为概念秩算子，其奇异值支配着从像素空间到类别logits的投影中概念级信号得以保留的程度，并将坍缩审计定义为保留的概念秩能量与总概念能量之比。在低标签情形下，概念标注稀缺且审计必须由少量标注样本估计，该上界预测审计可靠性会随概念秩谱和标签预算的函数而退化。本分析是一项理论和方法论贡献：推导出该上界，陈述坍缩审计可识别的条件，并在声明的方差结构（簇间SD = 6.0，残差SD = 4.5，组内相关系数 = 0.64）下刻画审计作为决策程序的操作特征。在2000个设计单元中，针对簇数为8、16、24和32以及效应量为3、5和8个百分点的情形估计了功效；在80%功效下达到5个百分点需要52个簇。该上界将概念审计重新框定为容量检验而非保真度检验，并对可解释病害分类器应如何支持小农杀菌剂选择性决策具有启示意义。","Zenodo (CERN European Organization for Nuclear Research)","2026-10-21T00:00:00Z",69,{"impact":113,"substance":259,"depth":19,"authority":17,"freshness":13,"relevant":21,"comment":260},21,"理论方法论文，提出可解释性容量上界并给出审计功效分析，对智慧农业病害识别可信度有参考价值，但属小众方法学研究、缺乏田间实证，暂不宜进入每日精选。",[262],{"name":255,"url":252},[188,26,264,265,266],"小农户","植物病害识别","可解释性",{"code":268,"slug":269,"name":270,"path":271,"parent_code":272},"G2","farm-services","社会化服务与规模经营","\u002Ftopic\u002Frural-services\u002Ffarm-services","G",[274],{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},{"status":39,"locked":14,"version":40,"method":41,"evidence":276},{"candidates":277,"agricultural_evidence":46},[278,283,288],{"code":268,"score":17,"supported":46,"title":279,"abstract":281,"tags":282},[280],"smallholder",[280],[264],{"code":32,"score":13,"supported":46,"title":284,"abstract":286,"tags":287},[285],"computer vision",[],[26,265],{"code":61,"score":21,"supported":14,"title":289,"abstract":290,"tags":291},[],[],[188],[293,294],"农业人工智能 植物病害识别 可解释性 智慧农业","农业人工智能 植物病害识别","农业人工智能植物病害识别可解释性智慧农业-5592","10.5281\u002Fzenodo.23545193",{"doi":296,"openalex_id":298,"authors":299,"venue":255,"cited_by_count":77,"oa_url":252,"card":306,"direction":98,"ingested_from":99},"W7223137323",[300,302,304],{"name":301,"orcid":9},"Akshay Ramgoolam",{"name":303,"orcid":9},"Priya Bundhoo Appadoo",{"name":305,"orcid":9},"Jean-Noël LEspérance",{"tldr":307,"method":308,"finding":309,"direction":98,"opportunity":310},"提出可解释性容量上界，评估低标签下概念秩崩溃审计的可靠性。","将分类器建模为概念秩算子，推导容量上界并做2000格功效模拟。","审计可靠性随概念秩谱与标签预算下降；5个百分点效应需52个聚类达80%功效。","可将容量上界用于小农户杀菌剂选择性决策的可解释模型审计与标签预算设计。","2026-10-22T11:21:10.425290Z",{"id":313,"title":314,"url":315,"summary":316,"summary_zh":317,"content":9,"source_name":255,"source_url":315,"published_at":256,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":257,"score_detail":318,"sources":320,"tags":322,"primary_topic":326,"secondary_topics":332,"classification":334,"search_phrases":356,"slug":359,"view_count":77,"doi":360,"paper":361,"created_at":373},5591,"Hierarchical Conformal Risk Control for Artificial Intelligence Agriculture Irrigation Scheduling Under Unmodeled Infiltration: A Computer Safety-Guarantee Evaluation","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23543665","Irrigation scheduling algorithms that rely on soil-moisture forecasts are vulnerable to unmodeled infiltration, the process by which water bypasses the root zone through preferential pathways that resistance-based soil models cannot represent. The framework is evaluated through a replication-based design study that characterises the operating characteristics of the before-after-control-impact estimator used to audit scheduler performance, under a declared variance structure with between-cluster standard deviation 6.0, residual standard deviation 4.5 and intraclass correlation 0.64. Power was estimated across 2000 design cells spanning cluster counts of 8, 16, 24 and 32 and effect sizes of 3, 5 and 8 percentage points. The analysis indicates that 54 clusters are required to detect a 5-percentage-point effect at 80% power, and that the minimum detectable effect declines steeply as cluster count rises from 8 to 32. The contribution is a formal safety guarantee for irrigation decision support that remains valid when the infiltration model is wrong, together with an audit design that makes the guarantee empirically checkable.","依赖土壤湿度预报的灌溉调度算法容易受到未建模入渗的影响，即水分通过优先路径绕过根区的过程，而基于阻力的土壤模型无法表征这一过程。该框架通过一项基于重复的设计研究进行评估，该研究刻画了用于审计调度器性能的前后对照影响估计量的运行特性，其所依据的声明方差结构为：组间标准差6.0、残差标准差4.5、组内相关系数0.64。功效估计覆盖2000个设计单元，涵盖聚类数8、16、24和32以及效应量3、5和8个百分点。分析表明，在80%功效下检测5个百分点效应需要54个聚类，且最小可检测效应随聚类数从8增至32而急剧下降。本文的贡献在于为灌溉决策支持提供了一种正式的安全保证，该保证在入渗模型错误时仍然有效，同时提出了一种使该保证可经实证检验的审计设计。",{"impact":113,"substance":18,"depth":19,"authority":17,"freshness":20,"relevant":21,"comment":319},"该研究提出在入渗模型失配条件下仍有效的灌溉决策安全保证框架，并给出可审计的试验设计，方法新颖、结论可靠，但属细分方法学研究，产业影响有限，适合作为专业精选而非头条。",[321],{"name":255,"url":315},[188,26,323,324,325],"灌溉决策","安全保证","试验设计",{"code":327,"slug":328,"name":329,"path":330,"parent_code":331},"H3","irrigation","水资源与灌溉","\u002Ftopic\u002Fenvironment\u002Firrigation","H",[333],{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},{"status":39,"locked":14,"version":40,"method":41,"evidence":335},{"candidates":336,"agricultural_evidence":46},[337,342,348,352],{"code":327,"score":45,"supported":46,"title":338,"abstract":339,"tags":341},[328],[340,328],"灌溉",[323],{"code":32,"score":17,"supported":46,"title":343,"abstract":345,"tags":347},[344],"artificial intelligence",[346],"decision support",[26],{"code":53,"score":54,"supported":46,"title":349,"abstract":350,"tags":351},[],[57,58],[],{"code":61,"score":21,"supported":14,"title":353,"abstract":354,"tags":355},[],[],[188],[357,358],"灌溉调度 安全保证","农业人工智能 安全保证 智慧农业 灌溉决策","灌溉调度安全保证-5591","10.5281\u002Fzenodo.23543665",{"doi":360,"openalex_id":362,"authors":363,"venue":255,"cited_by_count":77,"oa_url":315,"card":368,"direction":98,"ingested_from":99},"W7223131205",[364,366],{"name":365,"orcid":9},"Nassima Benkhelifa",{"name":367,"orcid":9},"Yacine Aït Ahmed",{"tldr":369,"method":370,"finding":371,"direction":98,"opportunity":372},"提出分层保形风险控制，为未建模入渗下的灌溉调度提供安全保证。","分层保形风险控制与基于重复设计的BACI审计，2000个设计单元功效分析。","检测5个百分点效应需54个集群，最小可检测效应随集群数增加陡降。","可将保形风险控制扩展到其他未建模扰动下的农业决策安全保证与审计设计。","2026-10-22T11:21:10.283561Z",{"id":375,"title":376,"url":377,"summary":378,"summary_zh":379,"content":9,"source_name":255,"source_url":377,"published_at":256,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":380,"score_detail":381,"sources":383,"tags":385,"primary_topic":388,"secondary_topics":389,"classification":391,"search_phrases":406,"slug":409,"view_count":77,"doi":410,"paper":411,"created_at":425},5590,"Evaluation of Concept-Probe Temporal-Stability Audits for Computer Vision Agriculture Disease Classifiers Under Seasonal Phenology Drift: An Interpretability Decay Bound for Smallholder Fungicide Timing","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23545234","Concept-probe audits are increasingly used to certify that a crop-disease classifier's internal representations track the pathological concepts a smallholder would act on, but a probe certified at one point in the season may silently decouple from those concepts as phenology shifts the visual statistics of the field. the present analysis treats that decoupling as a formal object. It defines a temporal-stability audit as a sequence of concept-probe accuracies measured across defined phenological windows, and it derives an interpretability decay bound: the rate at which probe accuracy can fall before the fungicide-timing decision it is meant to support becomes unreliable. The bound is expressed as a function of the probe's initial margin, the drift in concept prevalence between windows, and the classifier's sensitivity to the drifted features, and it yields a certified audit interval rather than a single-point certificate. The paper then characterises the operating characteristics of the audit design under a declared variance structure, using a replication-based evaluation of the before-after-control-impact estimator with between-cluster SD = 6.0, residual SD = 4.5 and intraclass correlation = 0.64. Power was estimated across 2000 design cells over cluster counts of 8, 16, 24 and 32 and effect sizes of 3, 5 and 8 percentage points; 54 clusters were required for 5 percentage points at 80% power. The contribution is a decision-theoretic account of when a concept-probe certificate expires, and a design rule for scheduling re-audits across the season.","概念探针审计（concept-probe audit）日益被用于验证作物病害分类器的内部表征是否追踪小农户会据以采取行动的病理学概念，但在生长季某一时点获得认证的探针，可能随着物候变化改变田间的视觉统计特征而悄然与这些概念脱钩。本分析将这种脱钩视为一个形式化对象。它将时间稳定性审计（temporal-stability audit）定义为在既定物候窗口内测量的一系列概念探针准确率，并推导出一个可解释性衰减界限（interpretability decay bound）：即探针准确率在其所支持的杀菌剂施用时机决策变得不可靠之前所能下降的速率。该界限表示为探针初始裕度、窗口间概念流行率漂移以及分类器对漂移特征敏感度的函数，并给出一个经认证的审计区间而非单点认证。随后，本文在声明的方差结构下刻画该审计设计的运行特征，采用基于重复的前后对照影响估计量评估，其中组间标准差 = 6.0、残差标准差 = 4.5、组内相关系数 = 0.64。功效在 2000 个设计单元上估计，聚类数取 8、16、24 和 32，效应量取 3、5 和 8 个百分点；在 80% 功效下达到 5 个百分点需要 54 个聚类。本文的贡献在于从决策论角度阐明概念探针认证何时失效，并给出在整个生长季内安排重新审计的设计规则。",70,{"impact":113,"substance":18,"depth":19,"authority":17,"freshness":13,"relevant":21,"comment":382},"该论文提出概念探针审计的时序稳定性框架与可解释性衰减界，方法新颖、统计设计扎实，但属小众方法论研究，产业影响有限，适合作为专业精选而非头条。",[384],{"name":255,"url":377},[188,26,264,386,387],"作物病害识别","模型审计",{"code":268,"slug":269,"name":270,"path":271,"parent_code":272},[390],{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},{"status":39,"locked":14,"version":40,"method":41,"evidence":392},{"candidates":393,"agricultural_evidence":46},[394,398,402],{"code":268,"score":45,"supported":46,"title":395,"abstract":396,"tags":397},[280],[264,280],[264],{"code":32,"score":13,"supported":46,"title":399,"abstract":400,"tags":401},[285],[],[386,26],{"code":61,"score":21,"supported":14,"title":403,"abstract":404,"tags":405},[],[],[188],[407,408],"概念探针 时序稳定性 作物病害分类器","杀菌剂 施药时机 可解释性衰减","概念探针时序稳定性作物病害分类器-5590","10.5281\u002Fzenodo.23545234",{"doi":410,"openalex_id":412,"authors":413,"venue":255,"cited_by_count":77,"oa_url":377,"card":420,"direction":98,"ingested_from":99},"W7223088030",[414,416,418],{"name":415,"orcid":9},"Abdi-Basid Idriss Farah",{"name":417,"orcid":9},"Amina Abdillahi Guedi",{"name":419,"orcid":9},"Moussa Omar Youssouf",{"tldr":421,"method":422,"finding":423,"direction":98,"opportunity":424},"提出概念探针时间稳定性审计与可解释性衰减界，判断何时需重新审计。","定义跨物候窗口探针精度序列，推导衰减界，用BACI复制评估统计功效。","5个百分点效应需54个聚类达80%功效，审计证书会随物候漂移过期。","可将衰减界与再审计调度规则嵌入小农户杀菌剂时机决策支持系统，并验证跨作物泛化。","2026-10-22T11:21:08.978348Z",{"id":427,"title":428,"url":429,"summary":430,"summary_zh":431,"content":9,"source_name":255,"source_url":429,"published_at":256,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":432,"score_detail":433,"sources":435,"tags":437,"primary_topic":440,"secondary_topics":441,"classification":442,"search_phrases":459,"slug":462,"view_count":77,"doi":463,"paper":464,"created_at":474},5589,"Conformal-Risk Evaluation for Artificial Intelligence Agriculture Spray-Nozzle Fault Detection Under Label-Delay Feedback: A Computer Coverage Audit","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23544107","Spray-nozzle fault detection in precision agriculture is a classification problem in which the label that confirms a fault arrives after the spray pass, so the detector must act on unlabelled evidence and its error rate must be certified before the label exists. Conformal prediction offers distribution-free coverage guarantees, but the guarantee is stated for exchangeable calibration and test data, an assumption that label-delayed feedback violates because the calibration set is always older than the deployment set. the present analysis formalises label-delay feedback as a distribution shift in the conformity score and derives a coverage bound for split-conformal fault detectors under that shift. The bound separates the nominal miscoverage rate into an exchangeable component and a delay-induced component that scales with the total variation distance between the calibration and deployment score distributions. A replication-based evaluation of the ['Intercept', 'exposure', 'group', 'severity'] estimator was conducted: 600 data sets of N = 500 were drawn from a declared data-generating process (intercept 12.0, exposure 0.42, group -3.1, severity 1.35, error SD 4.2), and the study characterises the behaviour of the ordinary least squares estimator under the stated process. The estimator recovered the declared coefficients with estimated bias 0.01122 for the intercept and -0.0002 for exposure, root mean squared error 0.9217 and 0.01614 respectively, and empirical coverage of the nominal 95% intervals of 1.0 for both parameters across 600 replications. The coverage audit shows that the delay-induced term, not the exchangeable term, dominates the risk of a conformal spray-nozzle detector, and that the audit procedure can certify a detector before field labels return.","在精准农业中，喷雾喷嘴故障检测是一个分类问题，其中确认故障的标签在喷雾作业结束后才到达，因此检测器必须基于未标记的证据采取行动，并且其错误率必须在标签存在之前得到认证。保形预测（conformal prediction）提供了无分布覆盖保证，但该保证是针对可交换的校准数据和测试数据而言的，而标签延迟反馈违反了这一假设，因为校准集总是比部署集更旧。本分析将标签延迟反馈形式化为一致性分数中的分布偏移，并推导了在该偏移下分裂保形故障检测器的覆盖界。该界将名义错误覆盖 rate 分解为可交换分量和延迟诱导分量，后者随校准与部署分数分布之间的全变差距离而缩放。对 ['Intercept', 'exposure', 'group', 'severity'] 估计器进行了基于重复的评估：从声明的数据生成过程（截距 12.0，暴露 0.42，分组 -3.1，严重程度 1.35，误差标准差 4.2）中抽取了 600 个 N = 500 的数据集，研究刻画了普通最小二乘估计器在所述过程下的行为。该估计器恢复了声明的系数，截距的估计偏差为 0.01122，暴露为 -0.0002，均方根误差分别为 0.9217 和 0.01614，两个参数在 600 次重复中名义 95% 区间的经验覆盖率为 1.0。覆盖审计表明，延迟诱导项而非可交换项主导了保形喷雾喷嘴检测器的风险，并且该审计程序可以在田间标签返回之前对检测器进行认证。",62,{"impact":113,"substance":19,"depth":182,"authority":20,"freshness":13,"relevant":21,"comment":434},"论文提出标签延迟下共形预测覆盖界，方法有新意但属预印本、应用面窄，可作为专业补充而非每日精选头条。",[436],{"name":255,"url":429},[188,26,438,439],"精准施药","病虫害监测",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":14,"version":40,"method":41,"evidence":443},{"candidates":444,"agricultural_evidence":46},[445,449,454],{"code":32,"score":20,"supported":46,"title":446,"abstract":447,"tags":448},[344],[],[26],{"code":61,"score":450,"supported":46,"title":451,"abstract":452,"tags":453},7,[],[64,65],[188],{"code":455,"score":21,"supported":14,"title":456,"abstract":457,"tags":458},"F4",[],[],[438],[460,461],"农业人工智能 病虫害监测 智慧农业 精准施药","农业人工智能 病虫害监测","农业人工智能病虫害监测智慧农业精准施药-5589","10.5281\u002Fzenodo.23544107",{"doi":463,"openalex_id":465,"authors":466,"venue":255,"cited_by_count":77,"oa_url":429,"card":469,"direction":98,"ingested_from":99},"W7223116087",[467],{"name":468,"orcid":9},"Achieng' Atim Okello",{"tldr":470,"method":471,"finding":472,"direction":98,"opportunity":473},"将标签延迟反馈建模为共形分数分布偏移，推导覆盖率界并审计喷雾喷嘴故障检测器。","分裂共形预测、总变差距离、600次重复模拟（N=500）评估OLS估计。","延迟诱导项而非可交换项主导检测器风险，审计可在田间标签返回前认证检测器。","可将该覆盖率审计框架扩展到其他标签延迟的农业传感器故障检测，并研究延迟分布估计误差对认证的影响。","2026-10-22T11:21:02.845718Z"]