[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-1213":3,"related-1213":78},{"id":4,"title":5,"url":6,"summary":7,"summary_zh":8,"content":8,"source_name":9,"source_url":6,"published_at":10,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":14,"score_detail":15,"sources":22,"tags":24,"primary_topic":8,"secondary_topics":29,"classification":30,"search_phrases":48,"slug":51,"view_count":19,"doi":52,"paper":53,"created_at":77},1213,"Predicting hairy vetch (Vicia villosa Roth) biomass across climatic gradients: A tool for sustainable cover crop management","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.agsy.2026.104947","Predicting hairy vetch (Vicia villosa Roth) biomass across climatic gradients: A tool for sustainable cover crop management。Agricultural Systems",null,"Agricultural Systems","2026-08-29T00:00:00Z","论文",10,false,72,{"impact":16,"substance":17,"depth":16,"authority":18,"freshness":19,"relevant":20,"comment":21},18,20,14,2,1,"研究提出覆盖作物生物量预测工具，对可持续农业管理有实质贡献，但时效性较低。",[23],{"name":9,"url":6},[25,26,27,28],"智慧农业","覆盖作物","气候梯度","生物量预测",[],{"status":31,"locked":13,"version":32,"method":33,"evidence":34},"unclassified","2026-10-22.2","rules",{"candidates":35,"agricultural_evidence":46,"reason":47},[36,41],{"code":37,"score":20,"supported":13,"title":38,"abstract":39,"tags":40},"A7",[],[],[26],{"code":42,"score":20,"supported":13,"title":43,"abstract":44,"tags":45},"I1",[],[],[25],true,"缺少足够的主题证据",[49,50],"生物量预测 智慧农业 气候梯度 覆盖作物","生物量预测 智慧农业","生物量预测智慧农业气候梯度覆盖作物-1213","10.1016\u002Fj.agsy.2026.104947",{"doi":52,"openalex_id":54,"authors":55,"venue":9,"cited_by_count":75,"oa_url":8,"card":8,"direction":8,"ingested_from":76},"W7204524489",[56,58,61,63,65,67,70,72],{"name":57,"orcid":8},"Juan P. Renzi",{"name":59,"orcid":60},"Petr Smýkal","https:\u002F\u002Forcid.org\u002F0000-0002-6117-8510",{"name":62,"orcid":8},"Ignacio Ducos",{"name":64,"orcid":8},"Aldana Alonso",{"name":66,"orcid":8},"Clara Llorens",{"name":68,"orcid":69},"G. Garcı́a","https:\u002F\u002Forcid.org\u002F0000-0003-4058-4808",{"name":71,"orcid":8},"Jone Irazusta",{"name":73,"orcid":74},"Jan Brus","https:\u002F\u002Forcid.org\u002F0000-0001-5613-9798",0,"openalex","2026-09-01T04:03:05.806088Z",{"total":79,"page":20,"page_size":79,"items":80},6,[81,180,259,342,407,468],{"id":82,"title":83,"url":84,"summary":85,"summary_zh":86,"content":8,"source_name":87,"source_url":84,"published_at":88,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":14,"score_detail":89,"sources":96,"tags":98,"primary_topic":102,"secondary_topics":108,"classification":115,"search_phrases":161,"slug":164,"view_count":20,"doi":165,"paper":166,"created_at":179},4896,"Site-Specific Precision Cover Cropping for Soil Health and Environmental Sustainability Using Remote Sensing and Multi-Criteria Analysis","https:\u002F\u002Fdoi.org\u002F10.21663\u002Feeg-d-26-00041","ABSTRACT Cover crops, which are plants or mixtures of plants grown primarily to enhance and protect soil health, are a key strategy in sustainable agriculture. Traditionally, they have been uniformly planted in a field typically during off-season when the main crops are not in competition. However, advancements in precision agriculture, particularly the integration of remote-sensing technology, demand more efficient seeding strategies based on field variability. The aims of this study were to assess the specific needs of a field in Aurora County, SD, and determine the type and optimal combinations of the cover crop mix. The cover crop species considered in this study included legumes, grasses, and brassicas. The remote-sensing data were extracted from Sentinel-2 satellite imagery. Soil samples were collected from 0 to 30 cm depth, and soil properties were measured. Additionally, topographic maps such as slope and aspect were extracted from light detection and ranging (LiDAR) data. The results of precision cover crop analyses showed that 6.5 percent (7.8 ha), 20.6 percent (24.7 ha), 38.1 percent (45.7 ha), 29.6 percent (35.5 ha), and 5.3 percent (6.3 ha) of the studied field should planted in the following mixtures of crops: rye, radish, and vetch; triticale, turnip, and vetch; triticale, radish, and clover; triticale, rapeseed, and pea; and barley, flax, and clover, respectively. The mixtures significantly improved soil properties by reducing compaction, increasing residue C\u002FN ratio, and enhancing weed suppression compared with the control. The resulting map should be interpreted as a decision-support recommendation rather than a fully validated field prescription.","摘要 覆盖作物（cover crops）是指主要为了提升和保护土壤健康而种植的植物或植物混合体，是可持续农业的一项关键策略。传统上，覆盖作物通常在非主作物竞争的季节以均匀方式播种于田间。然而，精准农业的进步，尤其是遥感技术的整合，要求基于田块变异性制定更高效的播种策略。本研究旨在评估南达科他州奥罗拉县某田块的具体需求，并确定覆盖作物混合物的类型及最优组合。本研究考虑的覆盖作物种类包括豆科、禾本科和十字花科植物。遥感数据提取自Sentinel-2卫星影像。土壤样品采集深度为0至30 cm，并测定了土壤性质。此外，从激光雷达（LiDAR）数据中提取了坡度和坡向等地形图。精准覆盖作物分析结果表明，研究田块中分别有6.5%（7.8 ha）、20.6%（24.7 ha）、38.1%（45.7 ha）、29.6%（35.5 ha）和5.3%（6.3 ha）的面积应种植以下作物混合物：黑麦、萝卜和野豌豆；小黑麦、芜菁和野豌豆；小黑麦、萝卜和三叶草；小黑麦、油菜和豌豆；以及大麦、亚麻和三叶草。与对照相比，这些混合物通过降低紧实度、提高残体碳氮比和增强杂草抑制效果，显著改善了土壤性质。所得图谱应被解读为决策支持建议，而非经过充分验证的田间处方。","Environmental and Engineering Geoscience","2026-10-13T00:00:00Z",{"impact":90,"substance":91,"depth":92,"authority":93,"freshness":94,"relevant":20,"comment":95},12,21,17,13,9,"该研究将遥感与多准则分析结合用于精准覆盖作物制图，方法新颖、数据具体，对农业信息化与可持续农业有参考价值，但属单点田间案例，影响范围有限。",[97],{"name":87,"url":84},[25,99,100,101,26],"精准农业","遥感","土壤健康",{"code":103,"slug":104,"name":105,"path":106,"parent_code":107},"H2","soil","土壤健康与改良","\u002Ftopic\u002Fenvironment\u002Fsoil","H",[109],{"code":110,"slug":111,"name":112,"path":113,"parent_code":114},"I3","remote-sensing","遥感与农情监测","\u002Ftopic\u002Fdigital-agriculture\u002Fremote-sensing","I",{"status":116,"locked":13,"version":32,"method":33,"evidence":117},"classified",{"candidates":118,"agricultural_evidence":46},[119,125,130,136,143,150,157],{"code":103,"score":120,"supported":46,"title":121,"abstract":122,"tags":124},15,[104],[123,104],"土壤",[101],{"code":110,"score":90,"supported":46,"title":126,"abstract":128,"tags":129},[127],"remote sensing",[100],[100],{"code":42,"score":131,"supported":46,"title":132,"abstract":133,"tags":135},8,[],[99,134],"precision agriculture",[25,99],{"code":137,"score":79,"supported":46,"title":138,"abstract":139,"tags":142},"A4",[],[140,141],"油菜","rapeseed",[],{"code":144,"score":79,"supported":46,"title":145,"abstract":146,"tags":149},"H8",[],[147,148],"可持续农业","sustainable agriculture",[],{"code":151,"score":152,"supported":46,"title":153,"abstract":154,"tags":156},"B1",3,[],[155],"萝卜",[],{"code":37,"score":20,"supported":13,"title":158,"abstract":159,"tags":160},[],[],[26],[162,163],"Sentinel-2 遥感 覆盖作物","精准覆盖作物 土壤健康","Sentinel-2遥感覆盖作物-4896","10.21663\u002Feeg-d-26-00041",{"doi":165,"openalex_id":167,"authors":168,"venue":87,"cited_by_count":75,"oa_url":8,"card":173,"direction":177,"ingested_from":76},"W7221669942",[169,171],{"name":170,"orcid":8},"Ali Mirzakhani Nafchi",{"name":172,"orcid":8},"Salman Mirzaee",{"tldr":174,"method":175,"finding":176,"direction":177,"opportunity":178},"利用遥感与多准则分析为南达科他州田块制定精准覆盖作物混播方案。","Sentinel-2影像、LiDAR地形与土壤采样，结合多准则分析。","田块被划分为5个管理区，各推荐不同覆盖作物混播，显著改善土壤紧实度、残体C\u002FN比和抑草效果。","农业遥感与作物表型","可验证该决策支持地图的田间效果，并融合实时物联网数据优化动态推荐。","2026-10-14T23:30:27.077312Z",{"id":181,"title":182,"url":183,"summary":184,"summary_zh":185,"content":8,"source_name":186,"source_url":183,"published_at":187,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":188,"score_detail":189,"sources":193,"tags":195,"primary_topic":199,"secondary_topics":200,"classification":201,"search_phrases":231,"slug":234,"view_count":20,"doi":235,"paper":236,"created_at":258},3521,"A practical model and online tool to predict cover crop nitrogen content and carbon-to-nitrogen ratio via remote sensing","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs11119-026-10434-6","Abstract Purpose Estimations of cover crop (CC) nitrogen (N) content and carbon-to-nitrogen (C:N) ratio are crucial for guiding N fertilizer recommendations for subsequent cash crops, yet prediction tools that can be easily adopted by practitioners are lacking. This study aims to (1) develop models to predict CC N content and C:N ratio using Normalized Difference Vegetation Index (NDVI) and Normalized Difference Red Edge Index (NDRE) from Uncrewed Aerial Vehicles (UAV) and satellite imagery, combined with growing degree days (GDD, base 5 °C) and CC species; and (2) integrate the models developed into an online application. Methods Over 260 CC biomass samples were collected from 10 Pennsylvania counties in the Chesapeake Bay watershed representing diverse species (triticale, cereal rye, wheat, brassicas, legumes, and mixtures). Four models were evaluated: (1) UAV NDVI + GDD + species, (2) UAV NDRE + GDD + species, (3) satellite NDVI + GDD + species, and (4) satellite NDRE + GDD + species. Results Random forest models (R² > 0.74 for N content and R² > 0.86 for C: N ratio) generally explained more variability than generalized additive models (R² > 0.70 for N content and R² > 0.85 for C: N ratio) based on 10-fold cross validation. GDD and vegetation indices were the strongest predictors for N content, while GDD and species group influenced C: N ratio the most. In the independent site-years validation, UAV-NDRE explained the most variability in CC N content (R² = 0.82), and both UAV-based models (R² = 0.52) explained more variability than the satellite models for C: N ratio prediction. Conclusion Random forest models accurately predicted CC N content and C: N ratio and were integrated into an online application to support precision agriculture practices.","摘要 目的 估算覆盖作物（cover crop，CC）氮（N）含量及碳氮比（C:N）对于指导后续经济作物的氮肥推荐至关重要，但目前缺乏便于从业者采用的预测工具。本研究旨在（1）利用无人机（Uncrewed Aerial Vehicles，UAV）和卫星影像获取的归一化差异植被指数（Normalized Difference Vegetation Index，NDVI）和归一化差异红边指数（Normalized Difference Red Edge Index，NDRE），结合生长度日（growing degree days，GDD，基准温度5 °C）和覆盖作物物种，开发预测覆盖作物氮含量和碳氮比的模型；（2）将所开发模型整合至在线应用程序中。方法 在切萨皮克湾流域宾夕法尼亚州10个县采集了260余份覆盖作物生物量样品，涵盖多种物种（小黑麦、黑麦、小麦、芸薹属、豆科及混播）。评估了四种模型：（1）无人机NDVI + GDD + 物种，（2）无人机NDRE + GDD + 物种，（3）卫星NDVI + GDD + 物种，（4）卫星NDRE + GDD + 物种。结果 基于10折交叉验证，随机森林模型（氮含量R² > 0.74，碳氮比R² > 0.86）总体上比广义加性模型（氮含量R² > 0.70，碳氮比R² > 0.85）解释了更多的变异。GDD和植被指数是氮含量的最强预测因子，而GDD和物种组对碳氮比影响最大。在独立站点-年份验证中，无人机NDRE对覆盖作物氮含量变异的解释能力最强（R² = 0.82），且两个基于无人机的模型（R² = 0.52）在碳氮比预测方面均比卫星模型解释了更多的变异。结论 随机森林模型准确预测了覆盖作物氮含量和碳氮比，并已整合至在线应用程序中以支持精准农业实践。","Precision Agriculture","2026-09-25T00:00:00Z",79,{"impact":190,"substance":191,"depth":16,"authority":18,"freshness":94,"relevant":20,"comment":192},16,22,"基于无人机与卫星遥感的覆盖作物氮素与碳氮比预测模型并落地在线工具，方法扎实、数据规模可观，对精准施肥有直接应用价值。",[194],{"name":186,"url":183},[25,196,26,197,198],"精准施肥","遥感监测","氮素管理",{"code":110,"slug":111,"name":112,"path":113,"parent_code":114},[],{"status":116,"locked":13,"version":32,"method":33,"evidence":202},{"candidates":203,"agricultural_evidence":46},[204,208,213,220,227],{"code":110,"score":94,"supported":46,"title":205,"abstract":206,"tags":207},[127],[],[197],{"code":42,"score":209,"supported":46,"title":210,"abstract":211,"tags":212},7,[],[99,134],[25],{"code":214,"score":79,"supported":46,"title":215,"abstract":216,"tags":219},"A2",[],[217,218],"小麦","wheat",[],{"code":221,"score":222,"supported":46,"title":223,"abstract":224,"tags":226},"F3",4,[],[225],"fertilizer",[196],{"code":37,"score":20,"supported":13,"title":228,"abstract":229,"tags":230},[],[],[26],[232,233],"覆盖作物 氮含量 遥感 预测","UAV NDRE 碳氮比 在线工具","覆盖作物氮含量遥感预测-3521","10.1007\u002Fs11119-026-10434-6",{"doi":235,"openalex_id":237,"authors":238,"venue":186,"cited_by_count":75,"oa_url":183,"card":253,"direction":177,"ingested_from":76},"W7214325666",[239,241,243,245,247,249,251],{"name":240,"orcid":8},"N. Kaur",{"name":242,"orcid":8},"C. White",{"name":244,"orcid":8},"A. Lefever",{"name":246,"orcid":8},"A. Thieme",{"name":248,"orcid":8},"J. Jennewein",{"name":250,"orcid":8},"WD. Hively",{"name":252,"orcid":8},"D. Carrijo",{"tldr":254,"method":255,"finding":256,"direction":177,"opportunity":257},"用无人机和卫星遥感结合积温与物种，构建覆盖作物氮含量及碳氮比预测模型并开发在线工具。","260个覆盖作物样本，随机森林与广义加性模型，UAV\u002F卫星NDVI、NDRE、G","随机森林预测精度高（N含量R²>0.74，C:N比R²>0.86），UAV-NDRE对N含量解释力最","可探索多源遥感融合与实时在线工具在精准氮肥推荐中的田间验证和推广。","2026-09-26T23:30:02.797286Z",{"id":260,"title":261,"url":262,"summary":263,"summary_zh":264,"content":8,"source_name":265,"source_url":262,"published_at":266,"category":11,"cover_url":8,"hotness":267,"is_selected":13,"score":268,"score_detail":269,"sources":273,"tags":278,"primary_topic":282,"secondary_topics":286,"classification":287,"search_phrases":314,"slug":317,"view_count":209,"doi":318,"paper":319,"created_at":341},3470,"AI-generated advice as a reinforcement layer in climate-smart agriculture","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.landusepol.2026.108336","Climate-smart agricultural (CSA) practices are central to food-system decarbonisation, yet adoption often falls short of farmers’ stated intentions, weakening the impact of incentives and extension under capacity constraints. We test whether spatially targeted, AI-generated advice can narrow this intention-action gap in a randomised field experiment with 1529 row crop farmers in Iowa, Illinois and Indiana during the cover crop decision window. Farmers assigned to receive four AI-generated emails were 4.45 %age points more likely to plant cover crops than controls (z = 2.81, p = 0.005), despite high baseline intentions in both groups. Effects operated on the extensive margin: there was no detectable change in the share of land planted among adopters. Survey responses indicate high engagement and a shift from untested optimism to more calibrated trust after exposure. Supervised AI advice can provide a low-cost, scalable complement to existing extension, improving follow-through and modestly expanding uptake without displacing human expertise.","气候智慧型农业（CSA）实践是食品系统脱碳的核心，然而在能力受限的情况下，农户的实际采用往往低于其声称的意愿，削弱了激励措施和推广服务的效果。我们在爱荷华州、伊利诺伊州和印第安纳州开展了一项随机田间试验，覆盖1529名大田作物种植户，在覆盖作物决策窗口期测试了空间靶向的AI生成建议能否缩小这一意愿—行动差距。被分配接收四封AI生成电子邮件的农户种植覆盖作物的概率比对照组高4.45个百分点（z = 2.81，p = 0.005），尽管两组基线意愿均较高。效应体现在广延边际上：采用者中种植土地比例未检测到显著变化。调查回复表明参与度较高，且在接触建议后，农户从未经检验的乐观转向更为校准的信任。有监督的AI建议可作为现有推广服务的低成本、可扩展补充，改善后续落实并适度扩大采用，而不会取代人类专业知识。","Land Use Policy","2026-09-24T00:00:00Z",25,87,{"impact":191,"substance":270,"depth":271,"authority":18,"freshness":94,"relevant":20,"comment":272},23,19,"随机对照试验证实AI生成建议可低成本缩小农户意愿与行动差距，对智慧农业推广具参考价值。",[274,275],{"name":265,"url":262},{"name":276,"url":277},"Apollo","https:\u002F\u002Fdoi.org\u002F10.17863\u002Fcam.134741",[25,279,280,26,281],"农业人工智能","农业技术推广","气候智慧农业",{"code":42,"slug":283,"name":284,"path":285,"parent_code":114},"smart-agriculture","智慧农业总论","\u002Ftopic\u002Fdigital-agriculture\u002Fsmart-agriculture",[],{"status":116,"locked":13,"version":32,"method":33,"evidence":288},{"candidates":289,"agricultural_evidence":46},[290,295,299,304,309],{"code":42,"score":94,"supported":46,"title":291,"abstract":293,"tags":294},[292],"smart agriculture",[],[25],{"code":37,"score":20,"supported":13,"title":296,"abstract":297,"tags":298},[],[],[26],{"code":300,"score":20,"supported":13,"title":301,"abstract":302,"tags":303},"G1",[],[],[280],{"code":305,"score":20,"supported":13,"title":306,"abstract":307,"tags":308},"H7",[],[],[281],{"code":310,"score":20,"supported":13,"title":311,"abstract":312,"tags":313},"I2",[],[],[279],[315,316],"AI生成建议 覆盖作物","爱荷华 伊利诺伊 印第安纳 覆盖作物","AI生成建议覆盖作物-3470","10.1016\u002Fj.landusepol.2026.108336",{"doi":318,"openalex_id":320,"authors":321,"venue":265,"cited_by_count":75,"oa_url":262,"card":334,"direction":340,"ingested_from":76},"W7214223143",[322,325,327,329,332],{"name":323,"orcid":324},"Callum Alexander","https:\u002F\u002Forcid.org\u002F0009-0007-5275-4583",{"name":326,"orcid":8},"Aiora Zabala",{"name":328,"orcid":8},"Andreas Kontoleon",{"name":330,"orcid":331},"Shalamar Armstrong","https:\u002F\u002Forcid.org\u002F0000-0002-1326-9936",{"name":333,"orcid":8},"Anuoluwa Sangotayo",{"tldr":335,"method":336,"finding":337,"direction":338,"opportunity":339},"随机试验检验AI生成建议能否缩小农户覆盖作物种植的意图-行动差距。","1529户美国中西部农户随机对照试验，四次AI生成邮件干预。","AI建议使覆盖作物种植率提高4.45个百分点，效果体现在是否采纳而非种植面积。","农业人工智能与决策模型","可探索AI建议与人工推广协同、长期持续效果及不同作物区域的异质性影响。","智慧农业 \u002F 农业物联网","2026-09-25T23:30:09.887867Z",{"id":343,"title":344,"url":345,"summary":346,"summary_zh":347,"content":8,"source_name":348,"source_url":345,"published_at":349,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":350,"score_detail":351,"sources":353,"tags":355,"primary_topic":359,"secondary_topics":365,"classification":370,"search_phrases":387,"slug":390,"view_count":75,"doi":391,"paper":392,"created_at":406},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":131,"substance":91,"depth":16,"authority":90,"freshness":12,"relevant":20,"comment":352},"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。",[354],{"name":348,"url":345},[25,279,356,357,358],"小农户","植物病害识别","可解释性",{"code":360,"slug":361,"name":362,"path":363,"parent_code":364},"G2","farm-services","社会化服务与规模经营","\u002Ftopic\u002Frural-services\u002Ffarm-services","G",[366],{"code":310,"slug":367,"name":368,"path":369,"parent_code":114},"agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai",{"status":116,"locked":13,"version":32,"method":33,"evidence":371},{"candidates":372,"agricultural_evidence":46},[373,378,383],{"code":360,"score":90,"supported":46,"title":374,"abstract":376,"tags":377},[375],"smallholder",[375],[356],{"code":310,"score":12,"supported":46,"title":379,"abstract":381,"tags":382},[380],"computer vision",[],[279,357],{"code":42,"score":20,"supported":13,"title":384,"abstract":385,"tags":386},[],[],[25],[388,389],"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":391,"openalex_id":393,"authors":394,"venue":348,"cited_by_count":75,"oa_url":345,"card":401,"direction":338,"ingested_from":76},"W7223137323",[395,397,399],{"name":396,"orcid":8},"Akshay Ramgoolam",{"name":398,"orcid":8},"Priya Bundhoo Appadoo",{"name":400,"orcid":8},"Jean-Noël LEspérance",{"tldr":402,"method":403,"finding":404,"direction":338,"opportunity":405},"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":408,"title":409,"url":410,"summary":411,"summary_zh":412,"content":8,"source_name":348,"source_url":410,"published_at":349,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":350,"score_detail":413,"sources":415,"tags":417,"primary_topic":421,"secondary_topics":426,"classification":428,"search_phrases":450,"slug":453,"view_count":75,"doi":454,"paper":455,"created_at":467},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":131,"substance":191,"depth":16,"authority":90,"freshness":94,"relevant":20,"comment":414},"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。",[416],{"name":348,"url":410},[25,279,418,419,420],"灌溉决策","安全保证","试验设计",{"code":422,"slug":423,"name":424,"path":425,"parent_code":107},"H3","irrigation","水资源与灌溉","\u002Ftopic\u002Fenvironment\u002Firrigation",[427],{"code":310,"slug":367,"name":368,"path":369,"parent_code":114},{"status":116,"locked":13,"version":32,"method":33,"evidence":429},{"candidates":430,"agricultural_evidence":46},[431,436,442,446],{"code":422,"score":120,"supported":46,"title":432,"abstract":433,"tags":435},[423],[434,423],"灌溉",[418],{"code":310,"score":90,"supported":46,"title":437,"abstract":439,"tags":441},[438],"artificial intelligence",[440],"decision support",[279],{"code":103,"score":79,"supported":46,"title":443,"abstract":444,"tags":445},[],[123,104],[],{"code":42,"score":20,"supported":13,"title":447,"abstract":448,"tags":449},[],[],[25],[451,452],"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":454,"openalex_id":456,"authors":457,"venue":348,"cited_by_count":75,"oa_url":410,"card":462,"direction":338,"ingested_from":76},"W7223131205",[458,460],{"name":459,"orcid":8},"Nassima Benkhelifa",{"name":461,"orcid":8},"Yacine Aït Ahmed",{"tldr":463,"method":464,"finding":465,"direction":338,"opportunity":466},"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":469,"title":470,"url":471,"summary":472,"summary_zh":473,"content":8,"source_name":348,"source_url":471,"published_at":349,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":474,"score_detail":475,"sources":477,"tags":479,"primary_topic":482,"secondary_topics":483,"classification":485,"search_phrases":500,"slug":503,"view_count":75,"doi":504,"paper":505,"created_at":519},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":131,"substance":191,"depth":16,"authority":90,"freshness":12,"relevant":20,"comment":476},"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。",[478],{"name":348,"url":471},[25,279,356,480,481],"作物病害识别","模型审计",{"code":360,"slug":361,"name":362,"path":363,"parent_code":364},[484],{"code":310,"slug":367,"name":368,"path":369,"parent_code":114},{"status":116,"locked":13,"version":32,"method":33,"evidence":486},{"candidates":487,"agricultural_evidence":46},[488,492,496],{"code":360,"score":120,"supported":46,"title":489,"abstract":490,"tags":491},[375],[356,375],[356],{"code":310,"score":12,"supported":46,"title":493,"abstract":494,"tags":495},[380],[],[480,279],{"code":42,"score":20,"supported":13,"title":497,"abstract":498,"tags":499},[],[],[25],[501,502],"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":504,"openalex_id":506,"authors":507,"venue":348,"cited_by_count":75,"oa_url":471,"card":514,"direction":338,"ingested_from":76},"W7223088030",[508,510,512],{"name":509,"orcid":8},"Abdi-Basid Idriss Farah",{"name":511,"orcid":8},"Amina Abdillahi Guedi",{"name":513,"orcid":8},"Moussa Omar Youssouf",{"tldr":515,"method":516,"finding":517,"direction":338,"opportunity":518},"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"]