[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-86":3,"related-86":75},{"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":31,"secondary_topics":37,"classification":38,"search_phrases":69,"slug":72,"view_count":73,"doi":8,"paper":8,"created_at":74},86,"浙江省农科院：牵头研发'浙里良田''知梅'两个农业智能体 发布于长三角智慧农业大会","https:\u002F\u002Fwww.toutiao.com\u002Farticle\u002F7667388882983551540\u002F","浙江省农业科学院牵头研发的'浙里良田''知梅'两个农业智能体在第四届长三角智慧农业发展大会上同时亮相。'浙里良田'以通义千问大模型为'中枢神经',实现高标准农田智能选址、智能审核、土壤精准诊断;'知梅'以数字人'小梅'为交互入口,围绕杨梅全产业链构建完整智能闭环。",null,"## 长三角智慧农业发展大会在杭州举行\n\n2026-07-28 09:44·[金台资讯](https:\u002F\u002Fwww.toutiao.com\u002Fc\u002Fuser\u002Ftoken\u002FMS4wLjABAAAAL8m7IqT5eR3-VgZjzlMhg66-z57FyRmydYIImMLAsQ4\u002F?source=tuwen_detail)\n\n7月26日至27日，以“数智兴农，共创未来”为主题的第四届长三角智慧农业发展大会在杭州举办。会上，浙江省农科院牵头研发的两个农业智能体“浙里良田”“知梅”同时亮相，来自全国农业科研院所、高校、龙头企业及人工智能领域近400位专家学者和企业家齐聚一堂，共谋人工智能赋能农业新质生产力发展。\n\n过去，高标准农田建设选址需要从多个部门收集国土空间规划、生态红线、土壤资源等各类资料开展人工研判，且项目材料审核靠人工逐页翻阅，耕地地力提升措施往往不精准。“浙里良田”智能体像一名全天候在线的农田管家——以通义千问大模型为“中枢神经”，接入卫星遥感影像、汇聚多源数据，叠加空间智能计算、检索增强生成等技术，将全省农田相关的海量数据真正“读懂、用活”，实现高标准农田建设项目智能选址、项目资料智能审核，以及土壤精准诊断与改良方案智能生成。\n\n在兰溪、仙居、青田等核心杨梅产区，一场产业数字化变革正悄然铺开。“知梅”智能体突破传统软件系统范式，以数字人“小梅”作为交互入口，依托剪枝修型等八项核心模型能力，搭建“问、管、卖”三大服务场景，面向种植农户、规模农场、龙头企业、政府产业管理部门等多元化主体，全面覆盖种植生产、基地管控、产销联动全流程，构建杨梅产业一体化完整智能闭环。\n\n大会设置两个平行专题分会场，分别围绕农业大数据与人工智能创新应用、农业智能装备研发与应用开展研讨。活动现场同步发布浙江省农业人工智能技术需求清单，面向长三角及全国科研单位、科技企业征集协同攻关伙伴。\n\n谢丹颖、王语嫣","浙江省农业科学院 2026-07-28","2026-07-28T01:44:00Z","报道",10,false,72,{"impact":17,"substance":18,"depth":19,"authority":20,"freshness":21,"relevant":22,"comment":23},22,18,15,12,5,1,"省级农科院牵头研发农业智能体，在长三角智慧农业大会发布，影响范围较广，内容有实质进展，信源权威，但时效性稍差。",[25],{"name":10,"url":6},[27,28,29,30],"智慧农业","浙江","农科院","智能体",{"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},[44,50,56,63],{"code":45,"score":20,"supported":46,"title":47,"abstract":48,"tags":49},"I1",true,[27],[27],[27],{"code":32,"score":51,"supported":46,"title":52,"abstract":53,"tags":55},4,[],[54],"大模型",[30],{"code":57,"score":58,"supported":46,"title":59,"abstract":60,"tags":62},"H1",3,[],[61],"高标准农田",[],{"code":64,"score":58,"supported":46,"title":65,"abstract":66,"tags":68},"H2",[],[67],"土壤",[],[70,71],"智慧农业 农科院 智能体 浙江","智慧农业 农科院","智慧农业农科院智能体浙江-86",14,"2026-07-31T00:00:54.641373Z",{"total":76,"page":22,"page_size":76,"items":77},6,[78,145,186,224,268,330],{"id":79,"title":80,"url":81,"summary":82,"summary_zh":83,"content":8,"source_name":84,"source_url":81,"published_at":85,"category":86,"cover_url":8,"hotness":13,"is_selected":14,"score":87,"score_detail":88,"sources":93,"tags":95,"primary_topic":99,"secondary_topics":100,"classification":101,"search_phrases":118,"slug":121,"view_count":22,"doi":122,"paper":123,"created_at":144},5194,"A Comprehensive Review on Multi-Modal Large Language Models and Agentic Architectures for Precision Agricultural Decision Support","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23411181","This article reviews recent developments in multi-modal large language models and agentic architectures for agricultural decision support. Forty-five representative studies published between 2023 and 2026 were selected across perception, language, orchestration, soil intelligence and governance, organised under a four-layer taxonomy, and classified against nine functional capabilities using a stated, reproducible procedure. The comparative analysis exposes a pronounced imbalance. Crop and disease perception is covered by 36 per cent of the selected studies, while dose-bearing prescription appears in 7 per cent, profit-aware recommendation in 4 per cent, and persistent per-field memory in a single study. Advisory tools in this literature therefore remain unable to exchange state: image-based scanners are rarely conditioned on measured soil chemistry, and crop-recommendation models optimise agronomic suitability without reference to farm economics. On this basis the article synthesises a conceptual reference architecture from the reviewed literature. It is a literature-derived synthesis rather than an implemented system, and calibrated abstention, evaluation protocols for longitudinal memory, and vernacular voice coverage emerge as the most urgent gaps.","本文综述了面向农业决策支持的多模态大语言模型与智能体架构的最新进展。按照可复现的既定流程，选取2023年至2026年间发表的45项代表性研究，涵盖感知、语言、编排、土壤智能与治理等方向，将其纳入一个四层分类体系，并依据九项功能能力进行分类。比较分析揭示出明显的不均衡：作物与病害感知在所选研究中占36%，而含剂量处方的研究仅占7%，利润感知推荐占4%，持久性田块记忆仅见于一项研究。因此，该领域的咨询工具仍无法交换状态：基于图像的扫描器很少以实测土壤化学状况为条件，作物推荐模型在优化农艺适宜性时也未考虑农场经济性。在此基础上，本文综合所综述文献提出一个概念性参考架构。该架构是文献衍生的综合框架，而非已实现的系统；校准弃权、纵向记忆的评估协议以及方言语音覆盖被列为最紧迫的空白。","Zenodo (CERN European Organization for Nuclear Research)","2026-10-16T00:00:00Z","论文",63,{"impact":20,"substance":18,"depth":89,"authority":90,"freshness":91,"relevant":22,"comment":92},16,8,9,"系统性综述揭示处方决策与农场经济维度严重缺位，对农业AI研发方向有参考价值，但属文献综述而非实证突破，影响范围有限。",[94],{"name":84,"url":81},[27,96,30,97,98],"农业人工智能","多模态大模型","农业决策支持",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":14,"version":40,"method":41,"evidence":102},{"candidates":103,"agricultural_evidence":46},[104,109,114],{"code":32,"score":73,"supported":46,"title":105,"abstract":107,"tags":108},[106],"decision support",[106],[96,97,30],{"code":64,"score":76,"supported":46,"title":110,"abstract":111,"tags":113},[],[67,112],"soil",[],{"code":45,"score":22,"supported":14,"title":115,"abstract":116,"tags":117},[],[],[27],[119,120],"多模态大模型 农业决策支持","农业人工智能 农业决策支持 多模态大模型 智慧农业","多模态大模型农业决策支持-5194","10.5281\u002Fzenodo.23411181",{"doi":122,"openalex_id":124,"authors":125,"venue":84,"cited_by_count":136,"oa_url":81,"card":137,"direction":141,"ingested_from":143},"W7222605322",[126,128,130,132,134],{"name":127,"orcid":8},"Tanmay Bhamre",{"name":129,"orcid":8},"Yash Barde",{"name":131,"orcid":8},"Devendra Yeole",{"name":133,"orcid":8},"Gajanan Chaudhary",{"name":135,"orcid":8},"Amarsinh Verpe",0,{"tldr":138,"method":139,"finding":140,"direction":141,"opportunity":142},"综述多模态大模型与智能体架构在精准农业决策支持中的应用与不足。","筛选2023-2026年45篇代表性研究，按四层分类与九项功能能力归类比较。","感知研究占36%，处方仅7%、利润推荐4%、田块记忆仅1篇，系统难以交换状态。","农业人工智能与决策模型","可研究跨模态状态交换、纵向记忆评测协议与方言语音覆盖，填补处方与经济决策空白。","openalex","2026-10-17T23:30:55.282694Z",{"id":146,"title":147,"url":148,"summary":149,"summary_zh":8,"content":150,"source_name":151,"source_url":8,"published_at":152,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":153,"score_detail":154,"sources":158,"tags":160,"primary_topic":163,"secondary_topics":164,"classification":165,"search_phrases":181,"slug":184,"view_count":90,"doi":8,"paper":8,"created_at":185},3938,"浙江省农科院农业人工智能创新中心算力服务平台建设（二期）项目中标","https:\u002F\u002Fsci-equip.net\u002Findex.php\u002Findex\u002Fbidder\u002Fdetail?id=246564","浙江省农业科学院农业人工智能创新中心算力服务平台建设（二期）项目中标结果公告显示，浙江移动数智科技有限公司以投标总价398.6万元中标，项目评审总得分85.1。代理机构为浙江省政府采购中心，合同信息详见公告附件。","## **关于浙江省农业科学院农业人工智能创新中心算力服务平台建设（二期）项目中标(成交)结果公告[浙江省政府采购中心]**\n\n浏览次数：39 发布日期：2026-09-28 12:33:32 来源：科器在线\n\n一、项目编号：330000263560010000421-ZZCG2026E-GK-147\n\n二、项目名称：浙江省农业科学院农业人工智能创新中心算力服务平台建设（二期）项目\n\n三、中标（成交）信息\n\n1.中标结果：\n\n**序号****中标（成交）金额(元)****中标供应商名称****中标供应商地址****评审总得分**\n1 投标总价：3986000（元）浙江移动数智科技有限公司 杭州市体育场路406号 85.1\n\n四、主要标的信息\n\n货物类主要标的信息：\n\n序号 标项名称 标的名称 品牌 规格型号 数量 单价(元)\n1 浙江省农业科学院农业人工智能创新中心算力服务平台建设（二期）项目 详见附件 详见附件 详见附件 详见附件 3986000\n\n五、评审专家（单一来源采购人员）名单\n\n龚建新，孙才俊（第1标项采购人代表），高夫迅，岳丹，智庆芳\n\n六、代理服务收费标准及金额：\n\n1.代理服务收费标准：无\n\n2.代理服务收费金额（元）：0\n\n七、公告期限\n\n自本公告发布之日起1个工作日。\n\n八、其他补充事宜\n\n1.各参加政府采购活动的供应商认为该中标\u002F成交结果和采购过程等使自己的权益受到损害的，可以自本公告期限届满之日（本公告发布之日后第2个工作日）起7个工作日内，以书面形式向采购人或受其委托的采购代理机构提出质疑。质疑供应商对采购人、采购代理机构的答复不满意或者采购人、采购代理机构未在规定的时间内作出答复的，可以在答复期满后十五个工作日内向同级政府采购监督管理部门投诉。质疑函范本、投诉书范本请到浙江政府采购网下载专区下载。\n\n2.其他事项：无\n\n九、对本次公告内容提出询问、质疑、投诉，请按以下方式联系\n\n1.采购人信息\n\n名 称：浙江省农业科学院（本级）\n\n地 址：杭州市石桥路198号\n\n传 真：\u002F\n\n项目联系人（询问）：孙才俊\n\n项目联系方式（询问）：13121117700\n\n质疑联系人：叶自然\n\n质疑联系方式：0571-88105537\n\n2.采购代理机构信息\n\n名 称：浙江省政府采购中心\n\n地 址：浙江省杭州市西湖区宝石一路3号\n\n传 真：\u002F\n\n项目联系人（询问）：吴皓哲\n\n项目联系方式（询问）：0571-88907791\n\n质疑联系人：张哲\n\n质疑联系方式：0571-88907707\n\n3.同级政府采购监督管理部门\n\n名 称：浙江省财政厅政府采购监管处、浙江省政府采购行政裁决服务中心（杭州）\n\n地 址：杭州市上城区清泰街549号城建综合大楼11楼\n\n传 真：\u002F\n\n联 系 人：匡老师\n\n监督投诉电话：0571-87227986（投诉专用电话）\n\n附件信息：\n\n*   [报价文件.pdf](https:\u002F\u002Fzcy-gov-open-doc.oss-cn-north-2-gov-1.aliyuncs.com\u002F1024FPA\u002F339900\u002F10016441811\u002F20269\u002Ff2a75c21-da03-4d50-8a08-1b6adce07406.pdf)\n\n351.2K\n\n*   [浙江省农业科学院农业人工智能创新中心算力服务平台建设（二期）项目采购文件.docx](https:\u002F\u002Fzcy-gov-open-doc.oss-cn-north-2-gov-1.aliyuncs.com\u002F1024FPA\u002F339900\u002F10013504574\u002F20268\u002Fb9d6d0aa-8538-406b-a100-7d56fa5799d9.docx)\n\n339.0K\n\n*   [关于符合本国产品标准的声明函](https:\u002F\u002Fzcy-gov-open-doc.oss-cn-north-2-gov-1.aliyuncs.com\u002F1024FPA\u002F20269\u002F769d3bc9-b2ea-433d-bf6e-a9de9f825f43.pdf)\n\n439.1K","浙江省政府采购中心·2026年09月29日","2026-09-29T00:00:00Z",50,{"impact":20,"substance":73,"depth":76,"authority":155,"freshness":156,"relevant":22,"comment":157},11,7,"省级农科院算力平台二期采购中标公告，属农业AI基础设施建设的实质进展，但内容为程序性公告，信息增量有限。",[159],{"name":151,"url":148},[27,96,28,161,162],"政府采购","算力平台",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":14,"version":40,"method":41,"evidence":166},{"candidates":167,"agricultural_evidence":46},[168,172,176],{"code":32,"score":20,"supported":46,"title":169,"abstract":170,"tags":171},[96],[96],[96],{"code":45,"score":22,"supported":14,"title":173,"abstract":174,"tags":175},[],[],[27],{"code":177,"score":22,"supported":14,"title":178,"abstract":179,"tags":180},"J1",[],[],[161],[182,183],"浙江省农科院 算力服务平台","浙江移动数智 农业人工智能","浙江省农科院算力服务平台-3938","2026-10-02T11:53:11.533911Z",{"id":187,"title":188,"url":189,"summary":190,"summary_zh":8,"content":191,"source_name":192,"source_url":8,"published_at":193,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":194,"score_detail":195,"sources":198,"tags":200,"primary_topic":202,"secondary_topics":203,"classification":204,"search_phrases":219,"slug":222,"view_count":156,"doi":8,"paper":8,"created_at":223},3577,"浙江省农业科学院农业人工智能创新中心算力服务平台建设（二期）项目中标公告——浙江移动数智科技中标，总金额398.6万元","https:\u002F\u002Fwww.ccgp.gov.cn\u002Fcggg\u002Fdfgg\u002Fzbgg\u002F202609\u002Ft20260925_27403210.htm","浙江省政府采购中心9月25日发布浙江省农业科学院农业人工智能创新中心算力服务平台建设（二期）项目中标公告，浙江移动数智科技有限公司以398.6万元中标，评审总得分85.1。评审专家包括龚建新、孙才俊（第1标项采购人代表）、高夫迅、岳丹、智庆芳。采购单位浙江省农业科学院位于杭州市石桥路198号。","##### 公告概要：\n\n**公告信息：**\n采购项目名称 浙江省农业科学院农业人工智能创新中心算力服务平台建设（二期）项目\n品目\n采购单位 浙江省农业科学院（本级）\n行政区域 浙江省 公告时间 2026年09月25日 14:52\n评审专家名单 龚建新，孙才俊（第1标项采购人代表），高夫迅，岳丹，智庆芳\n总中标金额￥398.600000 万元（人民币）\n**联系人及联系方式：**\n项目联系人 吴皓哲\n项目联系电话 0571-88907791\n采购单位 浙江省农业科学院（本级）\n采购单位地址 杭州市石桥路198号\n采购单位联系方式 13121117700\n代理机构名称 浙江省政府采购中心\n代理机构地址 浙江省杭州市西湖区宝石一路3号\n代理机构联系方式 0571-88907791\n\n一、项目编号：330000263560010000421-ZZCG2026E-GK-147\n\n二、项目名称：浙江省农业科学院农业人工智能创新中心算力服务平台建设（二期）项目\n\n三、中标（成交）信息\n\n1.中标结果：\n\n**序号****中标（成交）金额(元)****中标供应商名称****中标供应商地址****评审总得分**\n1 投标总价：3986000（元）浙江移动数智科技有限公司 杭州市体育场路406号 85.1\n\n四、主要标的信息\n\n货物类主要标的信息：\n\n序号 标项名称 标的名称 品牌 规格型号 数量 单价(元)\n1 浙江省农业科学院农业人工智能创新中心算力服务平台建设（二期）项目 详见附件 详见附件 详见附件 详见附件 3986000\n\n五、评审专家（单一来源采购人员）名单\n\n龚建新，孙才俊（第1标项采购人代表），高夫迅，岳丹，智庆芳\n\n六、代理服务收费标准及金额：\n\n1.代理服务收费标准：无\n\n2.代理服务收费金额（元）：0\n\n七、公告期限\n\n自本公告发布之日起1个工作日。\n\n八、其他补充事宜\n\n1.各参加政府采购活动的供应商认为该中标\u002F成交结果和采购过程等使自己的权益受到损害的，可以自本公告期限届满之日（本公告发布之日后第2个工作日）起7个工作日内，以书面形式向采购人或受其委托的采购代理机构提出质疑。质疑供应商对采购人、采购代理机构的答复不满意或者采购人、采购代理机构未在规定的时间内作出答复的，可以在答复期满后十五个工作日内向同级政府采购监督管理部门投诉。质疑函范本、投诉书范本请到浙江政府采购网下载专区下载。\n\n2.其他事项：无\n\n九、对本次公告内容提出询问、质疑、投诉，请按以下方式联系\n\n1.采购人信息\n\n名 称：浙江省农业科学院（本级）\n\n地 址：杭州市石桥路198号\n\n传 真：\u002F\n\n项目联系人（询问）：孙才俊\n\n项目联系方式（询问）：13121117700\n\n质疑联系人：叶自然\n\n质疑联系方式：0571-88105537\n\n2.采购代理机构信息\n\n名 称：浙江省政府采购中心\n\n地 址：浙江省杭州市西湖区宝石一路3号\n\n传 真：\u002F\n\n项目联系人（询问）：吴皓哲\n\n项目联系方式（询问）：0571-88907791\n\n质疑联系人：张哲\n\n质疑联系方式：0571-88907707\n\n3.同级政府采购监督管理部门\n\n名 称：浙江省财政厅政府采购监管处、浙江省政府采购行政裁决服务中心（杭州）\n\n地 址：杭州市上城区清泰街549号城建综合大楼11楼\n\n传 真：\u002F\n\n联 系 人：匡老师\n\n监督投诉电话：0571-87227986（投诉专用电话）\n\n附件信息：\n\n*   [报价文件.pdf](https:\u002F\u002Fzcy-gov-open-doc.oss-cn-north-2-gov-1.aliyuncs.com\u002F1024FPA\u002F339900\u002F10016441811\u002F20269\u002Ff2a75c21-da03-4d50-8a08-1b6adce07406.pdf)\n\n351.2K\n\n*   [浙江省农业科学院农业人工智能创新中心算力服务平台建设（二期）项目采购文件.docx](https:\u002F\u002Fzcy-gov-open-doc.oss-cn-north-2-gov-1.aliyuncs.com\u002F1024FPA\u002F339900\u002F10013504574\u002F20268\u002Fb9d6d0aa-8538-406b-a100-7d56fa5799d9.docx)\n\n339.0K\n\n*   [关于符合本国产品标准的声明函](https:\u002F\u002Fzcy-gov-open-doc.oss-cn-north-2-gov-1.aliyuncs.com\u002F1024FPA\u002F20269\u002F769d3bc9-b2ea-433d-bf6e-a9de9f825f43.pdf)\n\n439.1K","中国政府采购网 2026-09-25","2026-09-25T06:52:00Z",55,{"impact":20,"substance":73,"depth":90,"authority":196,"freshness":90,"relevant":22,"comment":197},13,"省级农科院AI算力平台二期采购中标公告，属农业信息化基础设施建设的实质性进展，但为常规政府采购结果公示，信息增量与专业深度有限。",[199],{"name":192,"url":189},[27,96,28,161,201],"算力服务",{"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":46},[207,211,215],{"code":32,"score":20,"supported":46,"title":208,"abstract":209,"tags":210},[96],[96],[96],{"code":45,"score":22,"supported":14,"title":212,"abstract":213,"tags":214},[],[],[27],{"code":177,"score":22,"supported":14,"title":216,"abstract":217,"tags":218},[],[],[161],[220,221],"浙江省农业科学院 算力服务平台","浙江移动数智科技 中标","浙江省农业科学院算力服务平台-3577","2026-09-27T00:05:15.284021Z",{"id":225,"title":226,"url":227,"summary":228,"summary_zh":8,"content":8,"source_name":229,"source_url":8,"published_at":230,"category":86,"cover_url":8,"hotness":13,"is_selected":14,"score":231,"score_detail":232,"sources":235,"tags":237,"primary_topic":240,"secondary_topics":241,"classification":242,"search_phrases":255,"slug":258,"view_count":51,"doi":8,"paper":259,"created_at":267},3439,"《Agentic Artificial Intelligence in Agriculture: A Systematic Mapping Review of Reported Architectures, Applications, Challenges, and Future Directions》","https:\u002F\u002Fwww.mdpi.com\u002F2227-7080\u002F14\u002F9\u002F591","作者按PRISMA 2020方案从4111项记录筛选至181项研究，对2020—2026年农业代理式人工智能文献做系统映射综述：领域跨度30多年但LLM子集非常年轻（2024年才出现、46\u002F47项发表于2025—2026年）；82%研究报告合作能力，计划和推理分别仅31%、记忆6%、反思4%；只有33项研究报告现场或实际部署，68项仍停留在概念性阶段，仅11项报告了一个季度以上的评估。","《Technologies》2026, 14(9), 591 \u002F MDPI","2026-09-22T00:00:00Z",80,{"impact":18,"substance":233,"depth":18,"authority":196,"freshness":90,"relevant":22,"comment":234},23,"基于PRISMA的农业代理式AI系统映射综述，量化揭示LLM应用年轻化与落地不足，信息增量与专业深度突出，值得进入每日精选。",[236],{"name":229,"url":227},[238,27,96,239,30],"数字农业","农业大模型",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":14,"version":40,"method":41,"evidence":243},{"candidates":244,"agricultural_evidence":46},[245,250],{"code":32,"score":155,"supported":46,"title":246,"abstract":248,"tags":249},[247],"artificial intelligence",[],[96,239,30],{"code":45,"score":251,"supported":14,"title":252,"abstract":253,"tags":254},2,[],[],[238,27],[256,257],"农业代理式人工智能 系统映射综述","农业人工智能 农业大模型 数字农业 智慧农业","农业代理式人工智能系统映射综述-3439",{"doi":8,"openalex_id":8,"authors":260,"venue":8,"cited_by_count":136,"oa_url":8,"card":261,"direction":141,"ingested_from":266},[],{"tldr":262,"method":263,"finding":264,"direction":141,"opportunity":265},"系统映射181项研究，梳理农业代理式AI的架构、应用、挑战与未来方向。","按PRISMA 2020筛选4111项记录至181项，做系统映射综述。","LLM代理2024年才出现，多具合作能力但规划、记忆、反思薄弱，实际部署少。","农业LLM代理的长期田间部署、记忆与反思机制及跨季度评估仍是明显空白。","agent","2026-09-25T00:09:33.995675Z",{"id":269,"title":270,"url":271,"summary":272,"summary_zh":273,"content":8,"source_name":84,"source_url":271,"published_at":274,"category":86,"cover_url":8,"hotness":13,"is_selected":14,"score":275,"score_detail":276,"sources":279,"tags":281,"primary_topic":285,"secondary_topics":291,"classification":293,"search_phrases":310,"slug":313,"view_count":136,"doi":314,"paper":315,"created_at":329},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。","2026-10-21T00:00:00Z",69,{"impact":90,"substance":277,"depth":18,"authority":20,"freshness":13,"relevant":22,"comment":278},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。",[280],{"name":84,"url":271},[27,96,282,283,284],"小农户","植物病害识别","可解释性",{"code":286,"slug":287,"name":288,"path":289,"parent_code":290},"G2","farm-services","社会化服务与规模经营","\u002Ftopic\u002Frural-services\u002Ffarm-services","G",[292],{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},{"status":39,"locked":14,"version":40,"method":41,"evidence":294},{"candidates":295,"agricultural_evidence":46},[296,301,306],{"code":286,"score":20,"supported":46,"title":297,"abstract":299,"tags":300},[298],"smallholder",[298],[282],{"code":32,"score":13,"supported":46,"title":302,"abstract":304,"tags":305},[303],"computer vision",[],[96,283],{"code":45,"score":22,"supported":14,"title":307,"abstract":308,"tags":309},[],[],[27],[311,312],"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":314,"openalex_id":316,"authors":317,"venue":84,"cited_by_count":136,"oa_url":271,"card":324,"direction":141,"ingested_from":143},"W7223137323",[318,320,322],{"name":319,"orcid":8},"Akshay Ramgoolam",{"name":321,"orcid":8},"Priya Bundhoo Appadoo",{"name":323,"orcid":8},"Jean-Noël LEspérance",{"tldr":325,"method":326,"finding":327,"direction":141,"opportunity":328},"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":331,"title":332,"url":333,"summary":334,"summary_zh":335,"content":8,"source_name":84,"source_url":333,"published_at":274,"category":86,"cover_url":8,"hotness":13,"is_selected":14,"score":275,"score_detail":336,"sources":338,"tags":340,"primary_topic":344,"secondary_topics":350,"classification":352,"search_phrases":372,"slug":375,"view_count":136,"doi":376,"paper":377,"created_at":389},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":90,"substance":17,"depth":18,"authority":20,"freshness":91,"relevant":22,"comment":337},"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。",[339],{"name":84,"url":333},[27,96,341,342,343],"灌溉决策","安全保证","试验设计",{"code":345,"slug":346,"name":347,"path":348,"parent_code":349},"H3","irrigation","水资源与灌溉","\u002Ftopic\u002Fenvironment\u002Firrigation","H",[351],{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},{"status":39,"locked":14,"version":40,"method":41,"evidence":353},{"candidates":354,"agricultural_evidence":46},[355,360,364,368],{"code":345,"score":19,"supported":46,"title":356,"abstract":357,"tags":359},[346],[358,346],"灌溉",[341],{"code":32,"score":20,"supported":46,"title":361,"abstract":362,"tags":363},[247],[106],[96],{"code":64,"score":76,"supported":46,"title":365,"abstract":366,"tags":367},[],[67,112],[],{"code":45,"score":22,"supported":14,"title":369,"abstract":370,"tags":371},[],[],[27],[373,374],"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":376,"openalex_id":378,"authors":379,"venue":84,"cited_by_count":136,"oa_url":333,"card":384,"direction":141,"ingested_from":143},"W7223131205",[380,382],{"name":381,"orcid":8},"Nassima Benkhelifa",{"name":383,"orcid":8},"Yacine Aït Ahmed",{"tldr":385,"method":386,"finding":387,"direction":141,"opportunity":388},"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"]