[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-2038":3,"related-2038":97},{"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":63,"slug":66,"view_count":67,"doi":68,"paper":69,"created_at":96},2038,"Trustworthy AutoML indicators for climate-smart agriculture: A sustainability-governance framework for decision support","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.indic.2026.101480","Trustworthy AutoML indicators for climate-smart agriculture: A sustainability-governance framework for decision support。Environmental and Sustainability Indicators","面向气候智慧型农业的可信自动机器学习指标：面向决策支持的可持续性治理框架。《环境与可持续性指标》",null,"Environmental and Sustainability Indicators","2026-09-09T00:00:00Z","论文",10,false,60,{"impact":17,"substance":18,"depth":19,"authority":20,"freshness":17,"relevant":21,"comment":22},8,16,15,13,1,"核心期刊论文，提出面向气候智慧型农业的可信AutoML指标与可持续治理决策框架，方法有新意但属学术探讨，公共影响有限。",[24],{"name":10,"url":6},[26,27,28,29,30],"智慧农业","农业人工智能","决策支持","气候智慧型农业","可持续治理",{"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":45},[44,51,58],{"code":32,"score":18,"supported":45,"title":46,"abstract":48,"tags":50},true,[47],"decision support",[49,47],"机器学习",[27,28],{"code":52,"score":53,"supported":45,"title":54,"abstract":56,"tags":57},"I1",12,[55],"smart agriculture",[55],[26],{"code":59,"score":21,"supported":14,"title":60,"abstract":61,"tags":62},"H7",[],[],[29],[64,65],"气候智慧型农业 农业人工智能 可持续治理 决策支持","气候智慧型农业 农业人工智能","气候智慧型农业农业人工智能可持续治理决策支持-2038",3,"10.1016\u002Fj.indic.2026.101480",{"doi":68,"openalex_id":70,"authors":71,"venue":10,"cited_by_count":87,"oa_url":6,"card":88,"direction":94,"ingested_from":95},"W7212009872",[72,75,77,79,81,84],{"name":73,"orcid":74},"Amar Singh","https:\u002F\u002Forcid.org\u002F0000-0002-5567-6521",{"name":76,"orcid":9},"Sukhbir Singh Walia",{"name":78,"orcid":9},"Ajay Singh",{"name":80,"orcid":9},"Shakti Kumar",{"name":82,"orcid":83},"Sahraoui Dhelim","https:\u002F\u002Forcid.org\u002F0000-0002-3620-1395",{"name":85,"orcid":86},"Chatter Singh","https:\u002F\u002Forcid.org\u002F0009-0005-6154-9237",0,{"tldr":89,"method":90,"finding":91,"direction":92,"opportunity":93},"提出面向气候智慧型农业的可信AutoML指标体系与可持续治理框架，用于决策支持。","构建可信AutoML指标与可持续治理框架，结合环境可持续性指标。","可信AutoML指标可支撑气候智慧型农业的可持续治理决策。","农业人工智能与决策模型","可探索可信AutoML指标在农业碳核算与气候适应决策中的实证验证与落地。","智慧农业 \u002F 农业物联网","openalex","2026-09-10T23:30:09.176914Z",{"total":98,"page":21,"page_size":98,"items":99},6,[100,186,257,318,373,436],{"id":101,"title":102,"url":103,"summary":104,"summary_zh":105,"content":9,"source_name":106,"source_url":103,"published_at":107,"category":12,"cover_url":9,"hotness":108,"is_selected":14,"score":109,"score_detail":110,"sources":115,"tags":119,"primary_topic":122,"secondary_topics":128,"classification":130,"search_phrases":170,"slug":173,"view_count":174,"doi":9,"paper":175,"created_at":185},4351,"Rhizosphere Microbiome Engineering for Climate-Smart Agriculture: From Synthetic Consortia to Precision Decision Support","https:\u002F\u002Fmel.cgiar.org\u002Freporting\u002Fdownloadmelspace\u002Fhash\u002F095138d10184dc7b696f79e27e716f84","Rhizosphere microbiome engineering is a promising approach that can enhance crop resilience and input use efficiency by redirecting plant–microbe–soil interactions toward predictable functions. Here, we review the mechanistic bases underlying rhizosphere assembly and stability, including root exudate-mediated selection, priority effects, keystone taxa, and metabolite-driven signaling, and connect these principles to proposed design rules for microbial inoculants. We present a generalizable Design–Build–Test–Learn (DBTL) framework for engineering synthetic microbial consortia, covering trait-to-module mapping (nutrient acquisition, phytohormone modulation, ACC deaminase activity, stress-protective metabolites, and biocontrol), compatibility screening, minimal yet robust community architectures, and iterative optimization driven by multi-omics and high-throughput phenotyping. Translation to field settings is framed as an engineering challenge defined by formulation and administration limitations, including carrier type, seed coating and encapsulation methods, shelf life, strain invasiveness, and permanence of colonization amid environmental diversity. We also summarize how integrative measurement pipelines (amplicon and shotgun sequencing, transcriptomics, metabolomics, and network or causal analyses) can advance microbiome studies from correlation to actionability. We describe how precision agriculture (sensors, remote sensing, and variable-rate inputs) and AI\u002FML (split-sample comparisons, transfer learning, and active learning) approaches can accelerate strain discovery, mixture optimization, and adaptive experimentation, driven by the need for stringent controls, metadata-rich reporting, and cross-site comparability. Use cases focus on stress conditions (drought, salinity, thermal extremes, and biotic stress) to demonstrate how microbial functions translate to agronomic outcomes and to highlight critical bottlenecks for reproducible, scalable microbiome products.","根际微生物组工程是一种有前景的方法，通过将植物–微生物–土壤相互作用导向可预测的功能，能够增强作物抗逆性和投入品利用效率。本文综述了根际组装与稳定性背后的机制基础，包括根系分泌物介导的选择、优先效应、关键类群和代谢物驱动的信号传导，并将这些原理与微生物接种剂拟议的设计规则相联系。我们提出了一个可推广的设计–构建–测试–学习（DBTL）框架，用于工程化合成微生物群落，涵盖性状到模块的映射（养分获取、植物激素调节、ACC脱氨酶活性、胁迫保护性代谢物和生物防治）、兼容性筛选、最小但稳健的群落架构，以及由多组学和高通量表型驱动的迭代优化。向田间环境的转化被界定为一项工程挑战，其受限于制剂和施用方面的制约，包括载体类型、种子包衣和包埋方法、货架期、菌株入侵性以及在环境多样性中定殖的持久性。我们还总结了整合性测量流程（扩增子和鸟枪法测序、转录组学、代谢组学以及网络或因果分析）如何推动微生物组研究从相关性走向可操作性。我们描述了精准农业（传感器、遥感和变量投入）以及人工智能\u002F机器学习（分样本比较、迁移学习和主动学习）方法如何加速菌株发现、混合优化和自适应实验，其驱动力在于对严格对照、富含元数据的报告和跨站点可比性的需求。应用案例聚焦于胁迫条件（干旱、盐渍、极端温度和生物胁迫），以展示微生物功能如何转化为农艺结果，并突出可重复、可规模化微生物组产品的关键瓶颈。","MELSpace (ICARDA (The International Center for Agricultural Research in Dry Areas))","2026-10-05T00:00:00Z",25,80,{"impact":111,"substance":112,"depth":111,"authority":113,"freshness":17,"relevant":21,"comment":114},18,22,14,"ICARDA 综述提出根际微生物组工程的 DBTL 框架并衔接精准农业与 AI，方法体系完整、结论可靠，对气候智慧型农业有参考价值，但属综述类论文、非全国性政策或重大突破。",[116,117],{"name":106,"url":103},{"name":106,"url":118},"https:\u002F\u002Fhdl.handle.net\u002F20.500.11766\u002F70900",[26,27,120,29,121],"根际微生物","合成菌群",{"code":123,"slug":124,"name":125,"path":126,"parent_code":127},"H2","soil","土壤健康与改良","\u002Ftopic\u002Fenvironment\u002Fsoil","H",[129],{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},{"status":39,"locked":14,"version":40,"method":41,"evidence":131},{"candidates":132,"agricultural_evidence":45},[133,140,146,150,157,164],{"code":123,"score":111,"supported":45,"title":134,"abstract":136,"tags":139},[135],"rhizosphere",[137,138,124,135],"土壤","根际",[120],{"code":52,"score":19,"supported":45,"title":141,"abstract":142,"tags":145},[55],[143,144],"精准农业","precision agriculture",[26],{"code":32,"score":53,"supported":45,"title":147,"abstract":148,"tags":149},[47],[49],[27],{"code":59,"score":151,"supported":45,"title":152,"abstract":153,"tags":156},7,[],[154,155],"干旱","drought",[29],{"code":158,"score":98,"supported":45,"title":159,"abstract":160,"tags":163},"I3",[],[161,162],"遥感","remote sensing",[],{"code":165,"score":67,"supported":45,"title":166,"abstract":167,"tags":169},"F4",[],[168],"生物防治",[],[171,172],"ICARDA 根际微生物 合成菌群","根际微生物 精准农业 决策支持","ICARDA根际微生物合成菌群-4351",4,{"doi":9,"openalex_id":176,"authors":177,"venue":106,"cited_by_count":87,"oa_url":103,"card":180,"direction":94,"ingested_from":95},"W7220812607",[178],{"name":179,"orcid":9},"Nourhan Fouad Mahmoud",{"tldr":181,"method":182,"finding":183,"direction":92,"opportunity":184},"综述根际微生物组工程，提出DBTL框架设计合成菌群，并融合精准农业与AI实现气候智慧型农业。","综述根际组装机制，提出DBTL框架，整合多组学、高通量表型、传感器与AI\u002FML方","微生物组工程需从相关走向可操作，精准农业与AI可加速菌株发现和菌群优化，但田间可重复性仍是瓶颈。","可研究AI驱动的根际微生物组设计-田间验证闭环，重点解决菌群定殖持久性与跨站点可迁移性。","2026-10-07T23:30:12.909703Z",{"id":187,"title":188,"url":189,"summary":190,"summary_zh":191,"content":9,"source_name":192,"source_url":189,"published_at":193,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":194,"score_detail":195,"sources":197,"tags":199,"primary_topic":202,"secondary_topics":203,"classification":204,"search_phrases":230,"slug":233,"view_count":234,"doi":235,"paper":236,"created_at":256},3881,"A Human-Centered Machine Learning Dashboard for Real-Time Smart Farming Decision Support","https:\u002F\u002Fdoi.org\u002F10.68012\u002Fshem.v2i2.302","The integration of machine learning (ML) into precision agriculture has advanced predictive accuracy for tasks such as crop monitoring, irrigation scheduling, and yield estimation. However, many existing IoT-based smart farming systems remain technically driven, offering limited attention to how farmers and field operators actually interact with and interpret model outputs. This gap often results in low adoption rates, as complex predictions fail to translate into actionable, trustworthy decisions for end users. This study proposes a human-centered machine learning dashboard designed to bridge predictive modeling with real-time, farmer-friendly decision support. The system integrates IoT sensor data streams with ML-based prediction modules, presented through an interactive, usability-tested interface that emphasizes clarity, accessibility, and interpretability over raw technical complexity. A working prototype was developed and evaluated using both quantitative usability metrics (e.g., task completion time, System Usability Scale scores) and qualitative feedback from target users to assess the dashboard's effectiveness in supporting field-level decisions. The proposed system architecture, dashboard interface, and underlying ML pipeline are implemented as open and reproducible source code, enabling replication and adaptation across similar smart farming contexts. Results indicate that the human-centered design significantly improves users' comprehension of model outputs and their confidence in acting on system recommendations, without compromising predictive performance. This research contributes a practical framework for designing machine intelligence systems that prioritize end-user experience alongside technical accuracy, offering implications for the broader development of human-centered AI in agricultural and IoT-driven decision support environments.","机器学习（ML）在精准农业中的融合提升了作物监测、灌溉调度和产量估算等任务的预测准确性。然而，许多现有的基于物联网的智慧农业系统仍然以技术驱动为主，对农民和田间操作人员实际如何与模型输出交互并加以解读关注有限。这一差距往往导致采用率偏低，因为复杂的预测无法转化为最终用户可操作、可信赖的决策。本研究提出了一种以人为中心的机器学习仪表盘，旨在将预测建模与实时、面向农民的决策支持相衔接。该系统将物联网传感器数据流与基于机器学习的预测模块相集成，并通过一个经过可用性测试的交互式界面呈现，强调清晰性、可及性和可解释性，而非原始的技术复杂性。我们开发了一个可运行的原型，并使用定量可用性指标（如任务完成时间、系统可用性量表评分）和来自目标用户的定性反馈进行评估，以考察该仪表盘在支持田间决策方面的有效性。所提出的系统架构、仪表盘界面和底层机器学习流水线均以开放且可复现的源代码形式实现，便于在类似的智慧农业场景中复制和适配。结果表明，以人为中心的设计显著提升了用户对模型输出的理解以及依据系统建议采取行动的信心，同时未损害预测性能。本研究为设计在技术准确性的同时优先考虑最终用户体验的机器智能系统提供了一个实用框架，对农业及物联网驱动的决策支持环境中以人为中心的人工智能的更广泛发展具有启示意义。","Smart Human-centered Emerging Research in Machine Intelligence","2026-09-30T00:00:00Z",64,{"impact":53,"substance":111,"depth":18,"authority":13,"freshness":17,"relevant":21,"comment":196},"提出以人为本的机器学习仪表盘，兼顾可用性与预测性能，并开源实现，对智慧农业落地有参考价值。",[198],{"name":192,"url":189},[26,27,200,28,201],"物联网","人机交互",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":14,"version":40,"method":41,"evidence":205},{"candidates":206,"agricultural_evidence":45},[207,213,217,223],{"code":32,"score":208,"supported":45,"title":209,"abstract":211,"tags":212},27,[210,47],"machine learning",[49,210,47],[27,28],{"code":52,"score":13,"supported":45,"title":214,"abstract":215,"tags":216},[],[26,143,144],[26],{"code":218,"score":151,"supported":45,"title":219,"abstract":220,"tags":222},"I5",[],[200,221],"IoT",[200],{"code":224,"score":98,"supported":45,"title":225,"abstract":226,"tags":229},"H3",[],[227,228],"灌溉","irrigation",[],[231,232],"农业人工智能 人机交互 决策支持 智慧农业","农业人工智能 人机交互","农业人工智能人机交互决策支持智慧农业-3881",2,"10.68012\u002Fshem.v2i2.302",{"doi":235,"openalex_id":237,"authors":238,"venue":192,"cited_by_count":87,"oa_url":189,"card":251,"direction":94,"ingested_from":95},"W7214922384",[239,242,244,246,248],{"name":240,"orcid":241},"Adam Faturahman","https:\u002F\u002Forcid.org\u002F0000-0001-9727-9092",{"name":243,"orcid":9},"James Williams",{"name":245,"orcid":9},"Evelyn Anderson",{"name":247,"orcid":9},"Goh Yao",{"name":249,"orcid":250},"Muhammad Rifat Syauqy","https:\u002F\u002Forcid.org\u002F0009-0001-4582-5693",{"tldr":252,"method":253,"finding":254,"direction":94,"opportunity":255},"提出以人为中心的机器学习仪表盘，融合物联网与预测模型，为农户提供实时可解释的决策支持。","集成物联网传感器数据流与机器学习预测模块，开发交互式仪表盘，并用SUS量表和用户","以人为中心的设计显著提升用户对模型输出的理解与采纳信心，且不牺牲预测性能。","可探索不同作物、地区和文化背景下农户对可解释AI界面的差异化需求与长期采纳效果。","2026-10-01T23:30:09.481397Z",{"id":258,"title":259,"url":260,"summary":261,"summary_zh":9,"content":9,"source_name":262,"source_url":260,"published_at":263,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":264,"score_detail":265,"sources":268,"tags":270,"primary_topic":273,"secondary_topics":279,"classification":281,"search_phrases":301,"slug":304,"view_count":67,"doi":305,"paper":306,"created_at":317},3807,"Reliability Auditing and Capacity-Constrained Review for Wheat-Yield Decision Support under Temporal and Spatial Shift","https:\u002F\u002Fdoi.org\u002F10.21203\u002Frs.3.rs-10698325\u002Fv1","Reliability Auditing and Capacity-Constrained Review for Wheat-Yield Decision Support under Temporal and Spatial Shift。Research Square","Research Square","2026-09-28T00:00:00Z",52,{"impact":17,"substance":113,"depth":19,"authority":98,"freshness":266,"relevant":21,"comment":267},9,"预印本论文，聚焦小麦产量决策支持系统的可靠性审计与时空迁移问题，方法有一定新意但尚未经同行评审，公共影响有限。",[269],{"name":262,"url":260},[26,27,28,271,272],"遥感监测","小麦产量",{"code":274,"slug":275,"name":276,"path":277,"parent_code":278},"A2","wheat","小麦","\u002Ftopic\u002Ffield-crops\u002Fwheat","A",[280],{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},{"status":39,"locked":14,"version":40,"method":41,"evidence":282},{"candidates":283,"agricultural_evidence":45},[284,288,293,297],{"code":32,"score":20,"supported":45,"title":285,"abstract":286,"tags":287},[47],[47],[27,28],{"code":274,"score":289,"supported":45,"title":290,"abstract":291,"tags":292},11,[275],[275],[],{"code":52,"score":21,"supported":14,"title":294,"abstract":295,"tags":296},[],[],[26],{"code":158,"score":21,"supported":14,"title":298,"abstract":299,"tags":300},[],[],[271],[302,303],"小麦产量 决策支持 遥感","Research Square 小麦 产量预测","小麦产量决策支持遥感-3807","10.21203\u002Frs.3.rs-10698325\u002Fv1",{"doi":305,"openalex_id":307,"authors":308,"venue":262,"cited_by_count":87,"oa_url":260,"card":9,"direction":92,"ingested_from":95},"W7214624863",[309,312,315],{"name":310,"orcid":311},"Xin Li","https:\u002F\u002Forcid.org\u002F0009-0005-0670-5006",{"name":313,"orcid":314},"Bojian Guo","https:\u002F\u002Forcid.org\u002F0009-0005-8353-2436",{"name":316,"orcid":9},"Asel Kartanova","2026-09-29T23:31:02.997580Z",{"id":319,"title":320,"url":321,"summary":322,"summary_zh":9,"content":9,"source_name":323,"source_url":9,"published_at":324,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":325,"score_detail":326,"sources":328,"tags":330,"primary_topic":332,"secondary_topics":333,"classification":334,"search_phrases":359,"slug":362,"view_count":234,"doi":9,"paper":363,"created_at":372},3246,"Geospatial Artificial Intelligence in Precision Agriculture: A Systematic Review（精准农业中的地理空间人工智能：系统综述）","https:\u002F\u002Fwww.mdpi.com\u002F3043-1204\u002F1\u002F1\u002F4","MDPI AI Precis. Agric. 1(1) 4 发表系统综述：综述2019-2026年Scopus、Web of Science及补充检索文献，考察精准农业中的应用、多模态数据集成与决策支持。综述表明研究范式从孤立的制图与预测转向集成工作流，整合卫星与无人机影像、田间传感器、气象、土壤与管理记录。报告效益包括产量预测改善、压力与病害更早检测、土壤制图更精细、灌溉与投入品更精准。然而，产量、环境和经济效益仍依赖当地条件和运行验证。核心发现是：地理参考数据本身不能确保真正的地理空间AI；可靠工作流必须解决空间依赖性、尺度不匹配、可迁移性和不确定性。可解释性和融入农场运营对可执行推荐同样重要。采用仍受数据互操作性、连接性、可负担性、隐私和技术能力限制。","MDPI AI in Precision Agriculture","2026-09-22T00:00:00Z",82,{"impact":111,"substance":112,"depth":111,"authority":113,"freshness":13,"relevant":21,"comment":327},"系统综述梳理2019-2026年GeoAI在精准农业的集成工作流与落地瓶颈，结论扎实、时效性强，对智慧农业技术路线有参考价值。",[329],{"name":323,"url":321},[26,27,161,28,331],"多模态数据",{"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":45},[337,342,347,351,355],{"code":52,"score":338,"supported":45,"title":339,"abstract":340,"tags":341},20,[143,144],[143],[26],{"code":32,"score":13,"supported":45,"title":343,"abstract":345,"tags":346},[344],"artificial intelligence",[],[27,28],{"code":123,"score":67,"supported":45,"title":348,"abstract":349,"tags":350},[],[137],[],{"code":224,"score":67,"supported":45,"title":352,"abstract":353,"tags":354},[],[227],[],{"code":158,"score":21,"supported":14,"title":356,"abstract":357,"tags":358},[],[],[161],[360,361],"GeoAI 精准农业 系统综述","地理空间人工智能 精准农业","GeoAI精准农业系统综述-3246",{"doi":9,"openalex_id":9,"authors":364,"venue":9,"cited_by_count":87,"oa_url":9,"card":365,"direction":369,"ingested_from":371},[],{"tldr":366,"method":367,"finding":368,"direction":369,"opportunity":370},"系统综述2019-2026年地理空间AI在精准农业的应用、多模态数据集成与决策支持。","系统综述Scopus、Web of Science等文献，分析多模态数据集成与决","地理参考数据不等于地理空间AI，可靠工作流须解决空间依赖、尺度、可迁移性与不确定性。","农业遥感与作物表型","可研究跨尺度空间依赖建模与可迁移性评估，提升模型在不同农场条件下的泛化能力。","agent","2026-09-23T00:04:33.172821Z",{"id":374,"title":375,"url":376,"summary":377,"summary_zh":378,"content":9,"source_name":379,"source_url":376,"published_at":380,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":381,"score_detail":382,"sources":385,"tags":387,"primary_topic":391,"secondary_topics":397,"classification":399,"search_phrases":416,"slug":419,"view_count":87,"doi":420,"paper":421,"created_at":435},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":17,"substance":383,"depth":111,"authority":53,"freshness":13,"relevant":21,"comment":384},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。",[386],{"name":379,"url":376},[26,27,388,389,390],"小农户","植物病害识别","可解释性",{"code":392,"slug":393,"name":394,"path":395,"parent_code":396},"G2","farm-services","社会化服务与规模经营","\u002Ftopic\u002Frural-services\u002Ffarm-services","G",[398],{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},{"status":39,"locked":14,"version":40,"method":41,"evidence":400},{"candidates":401,"agricultural_evidence":45},[402,407,412],{"code":392,"score":53,"supported":45,"title":403,"abstract":405,"tags":406},[404],"smallholder",[404],[388],{"code":32,"score":13,"supported":45,"title":408,"abstract":410,"tags":411},[409],"computer vision",[],[27,389],{"code":52,"score":21,"supported":14,"title":413,"abstract":414,"tags":415},[],[],[26],[417,418],"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":420,"openalex_id":422,"authors":423,"venue":379,"cited_by_count":87,"oa_url":376,"card":430,"direction":92,"ingested_from":95},"W7223137323",[424,426,428],{"name":425,"orcid":9},"Akshay Ramgoolam",{"name":427,"orcid":9},"Priya Bundhoo Appadoo",{"name":429,"orcid":9},"Jean-Noël LEspérance",{"tldr":431,"method":432,"finding":433,"direction":92,"opportunity":434},"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":437,"title":438,"url":439,"summary":440,"summary_zh":441,"content":9,"source_name":379,"source_url":439,"published_at":380,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":381,"score_detail":442,"sources":444,"tags":446,"primary_topic":450,"secondary_topics":453,"classification":455,"search_phrases":474,"slug":477,"view_count":87,"doi":478,"paper":479,"created_at":491},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":17,"substance":112,"depth":111,"authority":53,"freshness":266,"relevant":21,"comment":443},"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。",[445],{"name":379,"url":439},[26,27,447,448,449],"灌溉决策","安全保证","试验设计",{"code":224,"slug":228,"name":451,"path":452,"parent_code":127},"水资源与灌溉","\u002Ftopic\u002Fenvironment\u002Firrigation",[454],{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},{"status":39,"locked":14,"version":40,"method":41,"evidence":456},{"candidates":457,"agricultural_evidence":45},[458,462,466,470],{"code":224,"score":19,"supported":45,"title":459,"abstract":460,"tags":461},[228],[227,228],[447],{"code":32,"score":53,"supported":45,"title":463,"abstract":464,"tags":465},[344],[47],[27],{"code":123,"score":98,"supported":45,"title":467,"abstract":468,"tags":469},[],[137,124],[],{"code":52,"score":21,"supported":14,"title":471,"abstract":472,"tags":473},[],[],[26],[475,476],"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":478,"openalex_id":480,"authors":481,"venue":379,"cited_by_count":87,"oa_url":439,"card":486,"direction":92,"ingested_from":95},"W7223131205",[482,484],{"name":483,"orcid":9},"Nassima Benkhelifa",{"name":485,"orcid":9},"Yacine Aït Ahmed",{"tldr":487,"method":488,"finding":489,"direction":92,"opportunity":490},"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"]