[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-377":3,"related-377":78},{"id":4,"title":5,"url":6,"summary":7,"summary_zh":8,"content":9,"source_name":10,"source_url":8,"published_at":11,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":15,"score_detail":16,"sources":24,"tags":26,"primary_topic":30,"secondary_topics":36,"classification":42,"search_phrases":72,"slug":75,"view_count":76,"doi":8,"paper":8,"created_at":77},377,"中国小农户采用数字农业推广服务的驱动因素及福利效应——Double-Hurdle模型分析1475户果蔬种植户","https:\u002F\u002Fwww.ebiotrade.com\u002Fnewsf\u002F2026-8\u002F20260807103355465.htm","基于中国三个省份1475名果蔬种植户的微观调查数据,采用Double-Hurdle模型区分初始采纳与持续采纳,并识别不同驱动因素。研究进一步应用逆概率加权回归调整(IPWRA)评估DAES采纳对人均家庭净收入的福利效应。结果显示,初始采纳率为60.14%,但持续采纳率仅为24%,揭示显著的'采纳-持续采纳'差距。资源禀赋、风险偏好与社会网络等行为机制,以及数字可及性与政府信任等制度供给均显著促进这两个阶段,极端天气感知持续制约采纳行为。福利分析表明,采纳DAES使年度人均家庭净收入增加约12.6%。政策启示:缩小数字素养差距、开发气候适应型DAES、加强制度信任与社会网络。",null,"《NJAS: Impact in Agricultural and Life Sciences》：Drivers and welfare effects of adopting digital agricultural extension services among smallholder farmers in China\n\n_【字体： 大 中 小 】_ _时间：2026年08月07日_ _来源：NJAS: Impact in Agricultural and Life Sciences 1.6_\n\n**编辑推荐：**\n\n数字农业推广服务（DAES）旨在改善小农户的信息获取渠道，然而许多农户在初步尝试后停止使用的原因仍缺乏充分理解。基于中国三个省份1,475名果蔬种植户的微观调查数据，本研究采用Double-Hurdle模型区分了初始采纳与持续采纳，并识别了它们的不同驱动因素。\n\n数字农业推广服务（DAES）旨在改善小农户的信息获取渠道，然而许多农户在初步尝试后停止使用的原因仍缺乏充分理解。基于中国三个省份1,475名果蔬种植户的微观调查数据，本研究采用Double-Hurdle模型区分了初始采纳与持续采纳，并识别了它们的不同驱动因素。研究进一步应用逆概率加权回归调整（IPWRA）来评估DAES采纳对人均家庭净收入的福利效应。结果显示，初始采纳率为60.14%，但持续采纳率仅为24%，揭示了显著的“采纳-持续采纳”差距。资源禀赋、风险偏好与社会网络等行为机制，以及数字可及性与政府信任等制度供给均显著促进了这两个阶段，而感知的极端天气则持续制约采纳行为。福利分析表明，采纳DAES使年度人均家庭净收入增加约12.6%。通过将行为经济学与制度分析相结合，本研究阐明了农户从初始数字接入到持续使用的行为阶段性逻辑，并提出了数字素养支持、气候适应型服务以及制度信任重建的政策路径。主要政策启示包括：缩小数字素养差距以帮助农户从初始向持续采纳转变；开发在极端天气下保持有效的气候适应型DAES以克服持续使用的关键障碍；加强制度信任与社会网络以支持初始与持续采纳。\n\n**研究背景与意义**：农业推广服务是扩散农业技术、增强农村生计韧性及改善农户生产决策的重要制度安排。随着气候变化加剧和市场环境波动，小农户对及时、准确的农业信息需求不断累积，然而传统的线下推广模式在覆盖面、时效性和成本控制上存在明显局限，难以有效服务庞大的小农户群体。在此背景下，数字农业推广服务（DAES，依托移动互联网、数字服务平台等提供农业指导、市场信息和气象预警的服务）逐渐成为克服传统局限的有效工具。尽管中国农村互联网普及率显著提升，数字硬件约束得到缓解，但基础设施的改善并未带来DAES的常态化深度应用。许多农户在政策引导和邻里示范下进行初步接触后，仍停留在浅层试验阶段，难以将其纳入长期农业决策。现有的二元“采纳或不采纳”研究范式过度简化了农户使用DAES的内部过程，忽视了从初始接触到持续采纳的阶段异质性，且在福利效应评估中常面临样本自选择偏差。为解答农户从初始向持续采纳转型的驱动因素以及DAES常态化采纳能否有效提升农户收入的科学问题，研究人员开展了这项研究。该研究有助于厘清农户数字采纳的阶段性逻辑，为优化数字乡村建设中的服务供给、提升DAES包容性提供重要实证依据，论文发表在《NJAS: Impact in Agricultural and Life Sciences》。\n\n**关键技术方法**：研究人员主要利用2019年和2020年在山东、山西和陕西三省收集的1,475份果蔬种植户微观调查队列数据开展研究。在数据分析阶段，研究采用了Double-Hurdle模型将农户的采纳行为解构为初始采纳和持续采纳强度两个相互独立又关联的阶段，以克服传统Tobit模型在同质性假设上的缺陷；同时结合逆概率加权回归调整（IPWRA）构建反事实框架，以消除可观测协变量引起的自选择偏差并准确识别DAES对农户收入的净因果效应；此外，还运用了探索性因子分析（EFA）从浏览、下载、评价和分享四个维度构建了持续采纳指数，并采用Bootstrap法计算平均偏效应的标准误。\n\n**研究结果**：\n\n* **Double-Hurdle模型估计结果**：通过该模型分析发现，农户对DAES的采纳存在显著的阶段性差异。初始采纳率高达60.14%，但持续采纳率仅为24%，呈现出明显的“高接入、低粘性”特征。户主受教育程度对跨越初始采纳门槛具有显著正向作用，但对长期持续使用无显著影响；而耕地面积和人均家庭收入等资源禀赋在两个阶段均发挥显著正向促进作用。在行为机制方面，风险偏好和社会网络显著促进了初始与持续采纳；制度供给方面，数字网络设备获取便利性与对政府推广服务的满意度均显著正向影响两阶段决策。自然环境中，极端天气感知在两个阶段均产生显著负向影响，表明气候冲击并未激发信息避险需求，反而放大了农户对生产系统的整体不确定性感知。\n\n* **边际效应估计结果**：通过计算平均偏效应（APE）进一步量化了各因素的影响力。结果显示，数字网络设备获取便利性的平均边际效应最为突出（0.068），该变量每增加一个单位，初始采纳概率提升8.3%，持续采纳程度提升8.6%，是克服双重采纳障碍的关键基础设施。人均家庭收入和风险偏好的平均边际效应分别为0.032和0.038，对促进深度参与具有重要作用。户主性别、年龄、受教育程度、耕地面积、政府服务满意度及极端天气感知的整体平均边际效应未达统计显著水平，说明这些因素的影响存在阶段性局限。\n\n* **Heckman模型稳健性检验**：研究人员采用Heckman两阶段模型对基准结果进行检验。逆米尔斯比率（IMR）在1%水平上显著，证实了样本选择偏差的存在。在初始采纳方程中，核心变量的方向与显著性与Double-Hurdle模型高度一致，支持了核心结论的稳健性。在持续采纳方程中，因模型假设误差项联合正态分布，部分变量效应被选择修正项吸收，导致个别变量显著性发生变化，这恰好印证了Double-Hurdle模型在捕捉两阶段机制异质性上的适用性与必要性。\n\n* **DAES的福利效应**：通过IPWRA模型估计发现，DAES采纳对农户收入的平均处理效应（ATT）为0.126，在1%水平上显著。这表明在控制自选择偏差后，采纳DAES使农户年度人均家庭净收入增加约12.6%。通过平衡检验和多倾向得分匹配（PSM）方法的交叉验证，匹配后样本协变量偏差显著降低，不同匹配方法估计的增收效应稳定在34.7至38.9个百分点之间，证实了数字服务能实质性地改善农户家庭福利。\n\n**讨论与结论**：在讨论部分，研究人员指出，DAES的采纳并非简单的二元选择，而是从“初始采纳”向“持续采纳”演进的动态过滤过程，数字不平等已从传统的“接入鸿沟”转变为更深层的“使用鸿沟”。单一行为或制度视角无法解释这种阶段性差异，需将行为经济学与制度分析结合。行为机制上，农户受有限理性和损失厌恶影响，风险偏好通过降低心理不确定性促进持续采纳；制度维度上，熟人社会的网络扩散与政府公共机构的信任背书共同克服了农户的防御心理。极端天气感知的负向效应构成了“气候风险感知悖论”，表明若缺乏本地化的精准响应方案，外部冲击只会放大农户的失控感。研究最终得出以下结论：第一，DAES采纳表现出显著的阶段性“不连续性”，初始接入依靠行政动员，而持续采纳有赖于微观行为与宏观制度供给的匹配；第二，持续采纳取决于内部行为机制与外部制度供给的协同作用，极端天气感知持续抑制采纳，凸显了服务本地化适应的紧迫性；第三，DAES的收入红利并非数字接入的自动产物，而是源于生产实践中的持续嵌入。从初始接入到持续采纳的实质性转化，才是实现农业数字化转型包容性福利效应的核心前提。","NJAS: Impact in Agricultural and Life Sciences 2026-08-07","2026-08-07T00:00:00Z","论文",10,false,79,{"impact":17,"substance":18,"depth":19,"authority":20,"freshness":21,"relevant":22,"comment":23},18,24,20,14,3,1,"基于1475户样本的实证研究，揭示数字农业推广服务采纳的阶段性差异及增收效应，方法严谨，结论具有政策参考价值，但发表于一周前，时效性稍弱。",[25],{"name":10,"url":6},[27,28,29],"小农户","数字农业推广服务","Double-Hurdle NJAS",{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},"G2","farm-services","社会化服务与规模经营","\u002Ftopic\u002Frural-services\u002Ffarm-services","G",[37],{"code":38,"slug":39,"name":40,"path":41,"parent_code":35},"G1","extension-transfer","农技推广与成果转化","\u002Ftopic\u002Frural-services\u002Fextension-transfer",{"status":43,"locked":14,"version":44,"method":45,"evidence":46},"classified","2026-10-22.2","rules",{"candidates":47,"agricultural_evidence":50},[48,54,60,66],{"code":31,"score":49,"supported":50,"title":51,"abstract":52,"tags":53},9,true,[27],[],[27],{"code":38,"score":55,"supported":50,"title":56,"abstract":58,"tags":59},8,[57],"农业推广",[],[],{"code":61,"score":55,"supported":50,"title":62,"abstract":64,"tags":65},"I1",[63],"数字农业",[],[],{"code":67,"score":21,"supported":50,"title":68,"abstract":69,"tags":71},"H7",[],[70],"气候适应",[],[73,74],"数字农业推广服务 小农户 Double-Hurdle NJAS","数字农业推广服务 小农户","数字农业推广服务小农户Double-HurdleNJAS-377",2,"2026-08-10T23:54:46.300709Z",{"total":79,"page":22,"page_size":79,"items":80},6,[81,159,229,286,338,429],{"id":82,"title":83,"url":84,"summary":85,"summary_zh":86,"content":8,"source_name":87,"source_url":84,"published_at":88,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":89,"score_detail":90,"sources":94,"tags":96,"primary_topic":99,"secondary_topics":100,"classification":102,"search_phrases":131,"slug":134,"view_count":21,"doi":135,"paper":136,"created_at":158},3154,"The Role of Agricultural Extension Services in Promoting Climate-Smart Agriculture Adoption among Smallholder Farmers in Mwingi West sub-county Kitui County, Kenya","https:\u002F\u002Fdoi.org\u002F10.9734\u002Fajaees\u002F2026\u002Fv44i103022","Agricultural extension services provide smallholder farmers with knowledge, technical guidance, and advisory support that can facilitate the adoption of climate-smart agricultural (CSA) practices. This study examined the role of agricultural extension services in promoting CSA adoption among smallholder farmers in Mwingi West Sub-County, Kitui County, Kenya. A cross-sectional mixed-methods design was used. Quantitative data were collected from 200 randomly selected farmers through structured questionnaires, while qualitative evidence was obtained from 10 agricultural extension officers and 10 focus group discussions with farmer groups. Quantitative data were analysed using descriptive statistics and the chi-square test of independence. Significant associations were reported between CSA adoption and access to agricultural extension services, the source of extension information, type of extension support received, frequency of extension contact, perceived usefulness of extension services, CSA practices promoted through extension, and farmers’ perceived influence of extension services on adoption. Most respondents (82.0%) had access to extension services. Extension agents were the main information source (46.0%), and technical training was the most frequently reported support (38.0%). Conservation agriculture (30.0%), agroforestry (22.0%), and water harvesting (19.0%) were the leading practices promoted through extension activities. Nearly half of respondents (48.0%) reported that extension services had increased CSA adoption to a great extent. The findings indicate that accessible, practical, and regular extension support is closely associated with CSA adoption in the study area.","农业推广服务为小农户提供知识、技术指导和咨询服务，有助于促进气候智能型农业（CSA）实践的采纳。本研究考察了农业推广服务在促进肯尼亚基图伊县姆温吉西分区小农户采纳气候智能型农业实践中的作用。研究采用横断面混合方法设计。定量数据通过结构化问卷从200名随机抽取的农户中收集，定性证据则来自10名农业推广官员以及10次农户小组焦点小组讨论。定量数据采用描述性统计和卡方独立性检验进行分析。结果显示，气候智能型农业实践的采纳与以下因素存在显著关联：能否获得农业推广服务、推广信息来源、所获推广支持类型、推广接触频率、推广服务的感知有用性、通过推广所宣传的气候智能型农业实践，以及农户对推广服务在采纳过程中影响力的感知。大多数受访者（82.0%）能够获得推广服务。推广人员是主要的信息来源（46.0%），技术培训是最常报告的支持类型（38.0%）。通过推广活动宣传的主要实践包括保护性农业（30.0%）、农林业（22.0%）和雨水收集（19.0%）。近半数受访者（48.0%）表示推广服务在很大程度上提高了气候智能型农业实践的采纳率。研究结果表明，在研究区域内，可及、实用且定期的推广支持与气候智能型农业实践的采纳密切相关。","Asian Journal of Agricultural Extension Economics & Sociology","2026-09-22T00:00:00Z",65,{"impact":55,"substance":19,"depth":91,"authority":92,"freshness":13,"relevant":22,"comment":93},15,12,"基于肯尼亚小农户的实证研究，数据翔实但地域性强，对国内三农信息化参考价值有限，可作国际经验收录。",[95],{"name":87,"url":84},[97,27,98,57],"肯尼亚","气候智慧农业",{"code":38,"slug":39,"name":40,"path":41,"parent_code":35},[101],{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},{"status":43,"locked":14,"version":44,"method":45,"evidence":103},{"candidates":104,"agricultural_evidence":50},[105,111,116,121,127],{"code":38,"score":17,"supported":50,"title":106,"abstract":108,"tags":110},[107],"agricultural extension",[57,109,107],"技术培训",[57],{"code":31,"score":91,"supported":50,"title":112,"abstract":114,"tags":115},[113],"smallholder",[27,113],[27],{"code":61,"score":55,"supported":50,"title":117,"abstract":119,"tags":120},[118],"smart agriculture",[],[],{"code":122,"score":21,"supported":50,"title":123,"abstract":124,"tags":126},"H8",[],[125],"conservation agriculture",[],{"code":67,"score":22,"supported":14,"title":128,"abstract":129,"tags":130},[],[],[98],[132,133],"气候智慧农业 农业推广 小农户 肯尼亚","气候智慧农业 农业推广","气候智慧农业农业推广小农户肯尼亚-3154","10.9734\u002Fajaees\u002F2026\u002Fv44i103022",{"doi":135,"openalex_id":137,"authors":138,"venue":87,"cited_by_count":149,"oa_url":84,"card":150,"direction":156,"ingested_from":157},"W7213979623",[139,141,143,145,147],{"name":140,"orcid":8},"Simon Mbala",{"name":142,"orcid":8},"David Kawioo Mwanzia",{"name":144,"orcid":8},"Thomas Ekamais Akuja",{"name":146,"orcid":8},"Felix Ngunzo Kioli",{"name":148,"orcid":8},"Harrison Musyoka Maithya",0,{"tldr":151,"method":152,"finding":153,"direction":154,"opportunity":155},"研究肯尼亚小农户中农业推广服务对气候智慧型农业采纳的促进作用。","混合方法，200份农户问卷、10名推广员访谈和10组焦点小组，卡方检验。","推广服务可及性、信息源、支持类型和接触频率与CSA采纳显著相关，82%农户有推广服务。","数字乡村与农业信息化","可探索数字推广渠道（如手机短信、APP）如何替代或补充传统推广员，提升小农户CSA采纳率。","智慧农业 \u002F 农业物联网","openalex","2026-09-22T23:30:10.323072Z",{"id":160,"title":161,"url":162,"summary":163,"summary_zh":164,"content":8,"source_name":165,"source_url":162,"published_at":166,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":167,"score_detail":168,"sources":173,"tags":175,"primary_topic":179,"secondary_topics":185,"classification":188,"search_phrases":207,"slug":210,"view_count":76,"doi":211,"paper":212,"created_at":228},1747,"Artificial Intelligence and Large Language Models in Agricultural Extension: Critical Synthesis of Adoption, Efficacy, and Equity in Smallholder Advisory Systems","https:\u002F\u002Fdoi.org\u002F10.9734\u002Fair\u002F2026\u002Fv27i51723","Agricultural extension and advisory services are undergoing a rapid digital transformation driven by artificial intelligence (AI), machine learning, natural language processing, and, more recently, generative artificial intelligence and large language models (LLMs). These technologies have the potential to address persistent constraints in conventional agricultural extension, including limited adviser-to-farmer ratios, geographical isolation, inadequate localisation of information, and delays in the dissemination of climate, market, and agronomic information. LLM-based advisory systems are particularly significant because they can transform complex agricultural knowledge into conversational, multilingual, and potentially personalised guidance. However, their technical capability does not automatically translate into effective or equitable agricultural outcomes. This review synthesises emerging evidence on the adoption, efficacy, scalability, and equity implications of AI and LLM-based agricultural advisory systems, with particular emphasis on smallholder farmers in low- and middle-income countries. The literature indicates that AI-enabled advisory systems can improve access to timely information, facilitate question-answering, support crop and pest management, and extend the reach of human extension personnel. Emerging platforms such as Farmer.Chat demonstrate the feasibility of large-scale generative AI-enabled advisory delivery, while recent experiments with retrieval-augmented generation indicate promising pathways for improving factual grounding and contextual relevance. Nevertheless, evidence of direct impacts on farm productivity, profitability, resilience, and sustained behavioural change remains considerably weaker than evidence of technological feasibility and user engagement. Major limitations include hallucination, outdated or geographically inappropriate information, language and dialect constraints, weak interoperability with local agricultural databases, limited transparency, data governance concerns, and overreliance on automated recommendations. Equity is an equally important concern because digital access, literacy, gender, income, connectivity, and social norms influence who can benefit from AI-enabled services. The review argues that LLMs should therefore be positioned as augmentation technologies within pluralistic extension systems rather than as substitutes for human advisers. Future agricultural AI systems should combine retrieval from authoritative local knowledge bases, multimodal and multilingual interfaces, human oversight, participatory design, continuous evaluation, responsible data governance, and explicit equity metrics. The future value of AI in agricultural extension will ultimately depend not on model sophistication alone but on the quality of the socio-technical systems in which these technologies are embedded.","农业推广与咨询服务正在经历一场由人工智能（AI）、机器学习、自然语言处理以及近年来兴起的生成式人工智能和大语言模型（LLMs）驱动的快速数字化转型。这些技术有望解决传统农业推广中长期存在的制约因素，包括推广人员与农民比例有限、地理隔离、信息本地化不足，以及气候、市场和农艺信息传播的延迟。基于大语言模型的咨询系统尤为重要，因为它们能够将复杂的农业知识转化为对话式、多语言且可能实现个性化的指导。然而，其技术能力并不会自动转化为有效或公平的农业成果。本综述综合了关于基于AI和大语言模型的农业咨询系统在采用、有效性、可扩展性和公平性影响方面的新兴证据，特别关注低收入和中等收入国家的小农户。文献表明，AI赋能的咨询系统能够改善及时信息的获取，促进问答交互，支持作物与病虫害管理，并扩展人工推广人员的服务范围。诸如Farmer.Chat等新兴平台展示了大规模生成式AI赋能咨询服务的可行性，而近期关于检索增强生成的实验则表明，在改善事实依据和情境相关性方面存在有前景的路径。尽管如此，关于对农业生产率、盈利能力、韧性及持续性行为改变产生直接影响的证据，仍明显弱于关于技术可行性和用户参与度的证据。主要局限包括幻觉现象、信息过时或地理上不适用、语言和方言限制、与当地农业数据库的互操作性薄弱、透明度有限、数据治理问题，以及对自动化建议的过度依赖。公平性同样是一个重要关切，因为数字接入、素养、性别、收入、连通性和社会规范都会影响谁能够从AI赋能的服务中受益。本综述认为，大语言模型应被定位为多元推广体系中的增强技术，而非人类推广人员的替代品。未来的农业AI系统应将检索增强生成与结构化农业知识图谱、","Advances in Research","2026-09-04T00:00:00Z",82,{"impact":169,"substance":169,"depth":19,"authority":170,"freshness":171,"relevant":22,"comment":172},22,13,5,"系统综述AI与LLM在农技推广中的应用、效能与公平性，对智慧农业实践有重要参考价值。",[174],{"name":165,"url":162},[176,177,27,178],"农业人工智能","大语言模型","农技推广",{"code":180,"slug":181,"name":182,"path":183,"parent_code":184},"I2","agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai","I",[186,187],{"code":38,"slug":39,"name":40,"path":41,"parent_code":35},{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},{"status":43,"locked":14,"version":44,"method":45,"evidence":189},{"candidates":190,"agricultural_evidence":50},[191,199,203],{"code":180,"score":192,"supported":50,"title":193,"abstract":195,"tags":198},19,[194],"artificial intelligence",[196,197,194],"机器学习","machine learning",[176,177],{"code":38,"score":91,"supported":50,"title":200,"abstract":201,"tags":202},[107],[57,107],[178],{"code":31,"score":91,"supported":50,"title":204,"abstract":205,"tags":206},[113],[27,113],[27],[208,209],"农业人工智能 大语言模型 农技推广 小农户","农业人工智能 大语言模型","农业人工智能大语言模型农技推广小农户-1747","10.9734\u002Fair\u002F2026\u002Fv27i51723",{"doi":211,"openalex_id":213,"authors":214,"venue":165,"cited_by_count":149,"oa_url":162,"card":222,"direction":154,"ingested_from":157},"W7208785535",[215,218,220],{"name":216,"orcid":217},"Swati Sucharita","https:\u002F\u002Forcid.org\u002F0000-0003-3409-3295",{"name":219,"orcid":8},"Pratima Sachan",{"name":221,"orcid":8},"T. Sravan Kumar",{"tldr":223,"method":224,"finding":225,"direction":226,"opportunity":227},"综述AI与LLM在农业推广中的应用，评估其采纳、效果与公平性，尤其关注小农户。","系统综述新兴证据，分析AI\u002FLLM咨询系统的采用、效能、可扩展性及公平性。","AI可提升信息获取，但生产力影响证据弱，存在幻觉、公平性等问题，应作为增强而非替代。","农业人工智能与决策模型","研究空白：LLM在农业推广中的实际生产力影响与公平性机制，以及如何结合本地数据与人类顾问实现增强。","2026-09-05T23:30:36.654266Z",{"id":230,"title":231,"url":232,"summary":233,"summary_zh":234,"content":8,"source_name":235,"source_url":232,"published_at":236,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":237,"score_detail":238,"sources":241,"tags":243,"primary_topic":247,"secondary_topics":248,"classification":250,"search_phrases":266,"slug":269,"view_count":149,"doi":270,"paper":271,"created_at":285},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":55,"substance":239,"depth":17,"authority":92,"freshness":13,"relevant":22,"comment":240},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。",[242],{"name":235,"url":232},[244,176,27,245,246],"智慧农业","植物病害识别","可解释性",{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},[249],{"code":180,"slug":181,"name":182,"path":183,"parent_code":184},{"status":43,"locked":14,"version":44,"method":45,"evidence":251},{"candidates":252,"agricultural_evidence":50},[253,257,262],{"code":31,"score":92,"supported":50,"title":254,"abstract":255,"tags":256},[113],[113],[27],{"code":180,"score":13,"supported":50,"title":258,"abstract":260,"tags":261},[259],"computer vision",[],[176,245],{"code":61,"score":22,"supported":14,"title":263,"abstract":264,"tags":265},[],[],[244],[267,268],"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":270,"openalex_id":272,"authors":273,"venue":235,"cited_by_count":149,"oa_url":232,"card":280,"direction":226,"ingested_from":157},"W7223137323",[274,276,278],{"name":275,"orcid":8},"Akshay Ramgoolam",{"name":277,"orcid":8},"Priya Bundhoo Appadoo",{"name":279,"orcid":8},"Jean-Noël LEspérance",{"tldr":281,"method":282,"finding":283,"direction":226,"opportunity":284},"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":287,"title":288,"url":289,"summary":290,"summary_zh":291,"content":8,"source_name":235,"source_url":289,"published_at":236,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":292,"score_detail":293,"sources":295,"tags":297,"primary_topic":300,"secondary_topics":301,"classification":303,"search_phrases":318,"slug":321,"view_count":149,"doi":322,"paper":323,"created_at":337},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":55,"substance":169,"depth":17,"authority":92,"freshness":13,"relevant":22,"comment":294},"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。",[296],{"name":235,"url":289},[244,176,27,298,299],"作物病害识别","模型审计",{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},[302],{"code":180,"slug":181,"name":182,"path":183,"parent_code":184},{"status":43,"locked":14,"version":44,"method":45,"evidence":304},{"candidates":305,"agricultural_evidence":50},[306,310,314],{"code":31,"score":91,"supported":50,"title":307,"abstract":308,"tags":309},[113],[27,113],[27],{"code":180,"score":13,"supported":50,"title":311,"abstract":312,"tags":313},[259],[],[298,176],{"code":61,"score":22,"supported":14,"title":315,"abstract":316,"tags":317},[],[],[244],[319,320],"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":322,"openalex_id":324,"authors":325,"venue":235,"cited_by_count":149,"oa_url":289,"card":332,"direction":226,"ingested_from":157},"W7223088030",[326,328,330],{"name":327,"orcid":8},"Abdi-Basid Idriss Farah",{"name":329,"orcid":8},"Amina Abdillahi Guedi",{"name":331,"orcid":8},"Moussa Omar Youssouf",{"tldr":333,"method":334,"finding":335,"direction":226,"opportunity":336},"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":339,"title":340,"url":341,"summary":342,"summary_zh":343,"content":8,"source_name":344,"source_url":341,"published_at":236,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":345,"score_detail":346,"sources":349,"tags":351,"primary_topic":356,"secondary_topics":357,"classification":358,"search_phrases":394,"slug":397,"view_count":149,"doi":398,"paper":399,"created_at":428},5548,"Agroecological transition in olive based farming systems among smallholder farms in northwestern Tunisia","https:\u002F\u002Fdoi.org\u002F10.3389\u002Ffsufs.2026.1903591","Agricultural intensification has increased production and economic growth but has also generated significant environmental and social challenges that threaten long-term sustainability, highlighting the need for more sustainable farming models. Agroecology offers a promising pathway to enhance resilience and support inclusive rural development. In this context, Tunisia’s olive sector provides a relevant case for examining agroecological transitions in semi-arid smallholder systems. This study evaluates the degree of agroecological transition among smallholder olive farms in the Elles region of northwestern Tunisia using the FAO Tool for Agroecology Performance Evaluation (TAPE). Based on a sample of 60 farms, the analysis applies the Characterization of Agroecological Transition (CAET) index, grounded in the 10 Elements of Agroecology, to compare four production systems: (i) olive monoculture, (ii) diversified cropping, (iii) olive–livestock integration, and (iv) integrated olive–crop–livestock systems. Economic, social, environmental and institutional performance were explored using the TAPE Core Criteria of Performance (Step 2) and examined in relation to CAET scores. Results reveal a clear agroecological gradient, with diversified and integrated systems achieving markedly higher CAET scores (up to 52%) compared to monoculture systems (31%). Spearman correlations show strong associations between CAET and key dimensions, including synergy (ρ = 0.79), human and social values (ρ = 0.78), and diversity (ρ = 0.71). Several TAPE economic performance indicators, including monthly income, which ranged from approximately 1,349 TND in monoculture systems to 4,892 TND in diversified olive–crop–livestock systems, were positively associated with higher CAET scores, suggesting that agroecological transition may not necessarily imply a trade-off between productivity and sustainability. However, limited collective action, weak market integration, and insufficient extension services constrain broader adoption. These findings highlight the potential of diversification as a key lever for agroecological transition and underscore the importance of targeted policy support.","农业集约化提高了产量并促进了经济增长，但也产生了威胁长期可持续性的重大环境与社会挑战，凸显了对更可持续耕作模式的需求。农业生态学为增强韧性和支持包容性农村发展提供了一条有前景的路径。在此背景下，突尼斯橄榄产业为研究半干旱小农系统中的农业生态转型提供了相关案例。本研究利用联合国粮农组织农业生态性能评价工具（TAPE），评估突尼斯西北部埃勒斯地区小农橄榄农场的农业生态转型程度。基于60个农场的样本，分析采用以农业生态学十大要素为基础的农业生态转型特征（CAET）指数，比较四种生产系统：（i）橄榄单作，（ii）多样化种植，（iii）橄榄—畜牧结合，以及（iv）橄榄—作物—畜牧综合系统。研究利用TAPE核心性能标准（第二步）探讨经济、社会、环境和制度绩效，并将其与CAET得分进行关联分析。结果揭示了明显的农业生态梯度，多样化与综合系统的CAET得分（最高达52%）显著高于单作系统（31%）。斯皮尔曼相关性分析显示，CAET与关键维度之间存在强关联，包括协同性（ρ = 0.79）、人文与社会价值（ρ = 0.78）和多样性（ρ = 0.71）。若干TAPE经济绩效指标——包括月收入，其从单作系统的约1，349突尼斯第纳尔到橄榄—作物—畜牧多样化系统的4，892突尼斯第纳尔不等——与更高的CAET得分呈正相关，表明农业生态转型未必意味着生产力与可持续性之间的权衡。然而，集体行动有限、市场整合薄弱和推广服务不足制约了更广泛的采纳。这些发现凸显了多样化作为农业生态转型关键杠杆的潜力，并强调了有针对性的政策支持的重要性。","Frontiers in Sustainable Food Systems",71,{"impact":92,"substance":19,"depth":347,"authority":170,"freshness":49,"relevant":22,"comment":348},17,"基于FAO TAPE方法对突尼斯小农户橄榄系统农生态转型的实证研究，数据扎实、结论有新意，但属区域性案例，公共影响有限。",[350],{"name":344,"url":341},[27,352,353,354,355],"农业转型","生态农业","突尼斯","油橄榄",{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},[],{"status":43,"locked":14,"version":44,"method":45,"evidence":359},{"candidates":360,"agricultural_evidence":50},[361,365,372,378,385,390],{"code":31,"score":92,"supported":50,"title":362,"abstract":363,"tags":364},[113],[113],[27],{"code":366,"score":79,"supported":50,"title":367,"abstract":368,"tags":371},"C",[],[369,370],"畜牧","livestock",[],{"code":35,"score":79,"supported":50,"title":373,"abstract":374,"tags":377},[],[375,376],"农村发展","rural development",[],{"code":379,"score":79,"supported":50,"title":380,"abstract":381,"tags":384},"H6",[],[382,383],"农业生态","agroecology",[],{"code":67,"score":21,"supported":50,"title":386,"abstract":387,"tags":389},[],[388],"干旱",[],{"code":122,"score":22,"supported":14,"title":391,"abstract":392,"tags":393},[],[],[353],[395,396],"突尼斯 油橄榄 小农户","TAPE 农生态 评估","突尼斯油橄榄小农户-5548","10.3389\u002Ffsufs.2026.1903591",{"doi":398,"openalex_id":400,"authors":401,"venue":344,"cited_by_count":149,"oa_url":341,"card":422,"direction":426,"ingested_from":157},"W7223108121",[402,404,406,409,412,414,417,420],{"name":403,"orcid":8},"Rihab Mejri",{"name":405,"orcid":8},"Meriem Oueslati Zlaoui",{"name":407,"orcid":408},"Inès Gharbi","https:\u002F\u002Forcid.org\u002F0000-0003-2558-2134",{"name":410,"orcid":411},"Asma Souissi","https:\u002F\u002Forcid.org\u002F0000-0002-8680-5875",{"name":413,"orcid":8},"Ghada Ferchichi",{"name":415,"orcid":416},"Rania Soula","https:\u002F\u002Forcid.org\u002F0000-0002-6086-9176",{"name":418,"orcid":419},"Boubaker Dhehibi","https:\u002F\u002Forcid.org\u002F0000-0003-3854-6669",{"name":421,"orcid":8},"Mohamed Zied Dhraief",{"tldr":423,"method":424,"finding":425,"direction":426,"opportunity":427},"用FAO TAPE工具评估突尼斯西北小农橄榄农场生态农业转型程度。","FAO TAPE框架与CAET指数，调查60个农场比较四类种植系统。","多样化与综合系统CAET得分达52%，单作仅31%，且收入更高。","农业绿色发展与碳","可将TAPE与CAET指标数字化，构建小农生态转型监测与决策支持系统。","2026-10-22T11:20:15.716064Z",{"id":430,"title":431,"url":432,"summary":433,"summary_zh":8,"content":434,"source_name":435,"source_url":8,"published_at":236,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":167,"score_detail":436,"sources":438,"tags":440,"primary_topic":445,"secondary_topics":446,"classification":447,"search_phrases":459,"slug":462,"view_count":463,"doi":8,"paper":464,"created_at":472},5537,"农业产业园促进小农户融入数字化供应链：协同机制与制度逻辑","https:\u002F\u002Fkyy.nwafu.edu.cn\u002Fkjjz\u002Fxdkjjz\u002F557ffd662c8e4af98e46fbf90f35d91f.htm","西北农林科技大学阮俊虎教授团队联合德国哥廷根大学于晓华教授，以大荔县国家现代农业产业园等为典型案例，从交易成本理论出发围绕技术赋能、组织整合、规则治理与价值分配四个机制开展案例内分析与跨案例比较，揭示农业产业园作为中国特色制度平台促进小农户融入数字化供应链的深层逻辑。","近日，经济管理学院“数智农业运营管理团队”阮俊虎教授联合德国哥廷根大学于晓华教授在《中国农村经济》发表题为“农业产业园促进小农户融入数字化供应链：协同机制与制度逻辑”的研究论文，通过多案例研究揭示农业产业园促进小农户融入数字化供应链的协同机制，并阐释机制形成与运行的制度逻辑。博士研究生刘晨欢为第一作者，阮俊虎教授为通讯作者，王月博士和于晓华教授为合作学者。\n\n数字技术正深刻重构农产品供应链格局，大型电商平台等供应链核心主体加速构建覆盖生产源头至消费终端的数字化供应链体系，该体系不仅体现为数字技术在生产、销售、物流等环节的应用，更强调不同环节与经营主体之间的系统协同与数据贯通，小农户亟须有效融入数字化供应链以应对市场竞争格局的深刻变化。然而，“大国小农”国情农情下，小农户经营规模较小、土地细碎化程度较高，难以独立对接数字化、规模化的现代供应链体系。促进小农户深度融入数字化供应链，进而实现小农户和现代农业发展有机衔接，成为农业现代化进程中值得关注的重要议题。农业产业园作为区域农业发展的综合性制度平台，兼具资源集聚、组织协同与制度保障三重优势，能够为地方政府及园区经营管理主体协同整合制度资源并协调多元主体关系提供平台支撑，但将资源集聚、组织协同和制度供给优势转化为小农户融入数字化供应链的现实能力，其内在机理尚未得到系统性揭示。\n\n本文立足“农业产业园促进小农户融入数字化供应链的内在机理”这一科学问题，从交易成本理论出发，以大荔县国家现代农业产业园等为典型案例，围绕技术赋能、组织整合、规则治理和价值分配四个机制开展案例内分析与跨案例比较分析，揭示不同园区作用机制的共性、差异及情境适配特征，并从制度层面阐释农业产业园作为中国特色制度平台，促进小农户融入数字化供应链的深层逻辑。研究旨在为农业经营呈现小规模、分散化特征的国家和地区促进小农户和现代农业发展有机衔接提供理论参考与实践借鉴。\n\n该研究得到国家社会科学基金重大项目“我国农产品供应链数字化转型的发展战略与实施路径研究”（23&ZD123）的资助。\n\n论文链接：\n\n[https:\u002F\u002Fzgncjj.ajcass.com\u002F?jumpnotice=201606270007#\u002FissueDetail?contentId=123722&year=2026&issue=10&title=%E5%BD%93%E6%9C%9F%E7%9B%AE%E5%BD%95](https:\u002F\u002Fzgncjj.ajcass.com\u002F?jumpnotice=201606270007#\u002FissueDetail?contentId=123722&year=2026&issue=10&title=%E5%BD%93%E6%9C%9F%E7%9B%AE%E5%BD%95)\n\n编辑：张晴 终审：刘玉峰","中国农村经济 2026-10-21",{"impact":17,"substance":169,"depth":17,"authority":91,"freshness":49,"relevant":22,"comment":437},"核心期刊论文，多案例揭示农业产业园促进小农户融入数字化供应链的协同机制与制度逻辑，理论新意足、政策参考价值高，值得进入每日精选。",[439],{"name":435,"url":432},[441,27,442,443,444],"数字乡村","制度逻辑","数字化供应链","农业产业园",{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},[],{"status":43,"locked":14,"version":44,"method":45,"evidence":448},{"candidates":449,"agricultural_evidence":50},[450,454],{"code":31,"score":92,"supported":50,"title":451,"abstract":452,"tags":453},[27],[27],[27],{"code":455,"score":22,"supported":14,"title":456,"abstract":457,"tags":458},"I6",[],[],[441],[460,461],"农业产业园 小农户 数字化供应链","大荔县 国家现代农业产业园","农业产业园小农户数字化供应链-5537",7,{"doi":8,"openalex_id":8,"authors":465,"venue":8,"cited_by_count":149,"oa_url":8,"card":466,"direction":154,"ingested_from":471},[],{"tldr":467,"method":468,"finding":469,"direction":154,"opportunity":470},"以农业产业园为案例，揭示其促进小农户融入数字化供应链的机制与制度逻辑。","交易成本理论，大荔县等典型案例的案例内分析与跨案例比较。","技术赋能、组织整合、规则治理与价值分配四机制共同推动小农户融入数字化供应链。","可量化检验四机制对农户融入绩效的影响，并比较不同产业园模式的适用边界。","agent","2026-10-22T00:07:10.192700Z"]