[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-4369":3,"related-4369":87},{"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":9,"secondary_topics":31,"classification":32,"search_phrases":64,"slug":67,"view_count":68,"doi":9,"paper":69,"created_at":86},4369,"Adopsi Mobile Farming dan DSS dalam Pengambilan Keputusan Usaha Tani Sugihwaras","https:\u002F\u002Fjournal.udn.ac.id\u002Findex.php\u002Feduscotech\u002Farticle\u002Fview\u002F604","Digital agriculture can reduce information asymmetry and support farm decisions, yet adoption among smallholders depends on perceived benefits, usability, and enabling conditions. This study draft examines the adoption of mobile farming applications and decision support systems (DSS) for farm decision-making in Sugihwaras Village, Maospati, Magetan. An explanatory cross-sectional design is proposed. Because primary field data were not supplied, the present results use an explicitly simulated dataset of 96 farmers for manuscript development. Descriptive analysis and multiple regression indicate a relatively high mobile farming adoption score, moderate DSS use, and high perceived decision quality. Performance expectancy, facilitating conditions, social influence, and age are associated with mobile adoption, while mobile farming and DSS use are positively associated with decision quality. The findings framework highlights the need for user-friendly applications, reliable connectivity, farmer-group-based training, and extension support. Empirical field data are required before publication.","数字农业能够减少信息不对称并支持农场决策，但小农户的采用取决于其感知收益、可用性和使能条件。本研究初稿考察了在Magetan县Maospati镇Sugihwaras村，移动农业应用程序和决策支持系统（DSS）在农场决策中的采用情况。研究拟采用解释性横断面设计。由于未提供一手实地数据，当前结果使用了一个明确标注为模拟的96户农民数据集，用于稿件开发。描述性分析和多元回归表明，移动农业采用得分相对较高，DSS使用处于中等水平，感知决策质量较高。绩效期望、便利条件、社会影响和年龄与移动农业采用相关，而移动农业和DSS使用与决策质量呈正相关。研究结果框架强调了需要用户友好的应用程序、可靠的网络连接、基于农民小组的培训以及推广支持。在发表前仍需实证实地数据。",null,"Eduscotech","2026-10-05T00:00:00Z","论文",10,false,32,{"impact":17,"substance":18,"depth":19,"authority":20,"freshness":18,"relevant":21,"comment":22},6,8,7,3,1,"印尼小农户移动农业与DSS采纳研究，但数据为模拟、尚未发表，仅具参考价值。",[24],{"name":10,"url":6},[26,27,28,29,30],"数字乡村","智慧农业","决策支持系统","小农户","移动农业应用",[],{"status":33,"locked":14,"version":34,"method":35,"evidence":36},"needs_review","2026-10-22.2","rules",{"candidates":37,"agricultural_evidence":40,"reason":63},[38,46,52,58],{"code":39,"score":19,"supported":40,"title":41,"abstract":42,"tags":45},"I1",true,[],[43,44],"数字农业","digital agriculture",[27],{"code":47,"score":48,"supported":40,"title":49,"abstract":50,"tags":51},"G2",4,[],[29],[29],{"code":53,"score":48,"supported":40,"title":54,"abstract":55,"tags":57},"I2",[],[56],"decision support",[28],{"code":59,"score":21,"supported":14,"title":60,"abstract":61,"tags":62},"I6",[],[],[26],"多个产业证据相当，需要确认主归属",[65,66],"Sugihwaras 移动农业 采纳","Magetan 农户 决策支持系统","Sugihwaras移动农业采纳-4369",2,{"doi":9,"openalex_id":70,"authors":71,"venue":10,"cited_by_count":78,"oa_url":6,"card":79,"direction":83,"ingested_from":85},"W7220753702",[72,74,76],{"name":73,"orcid":9},"Paramita Wisnuwardhani",{"name":75,"orcid":9},"Eko Pujianto Eko Pujianto",{"name":77,"orcid":9},"Nunik Indarwati",0,{"tldr":80,"method":81,"finding":82,"direction":83,"opportunity":84},"基于模拟数据研究印尼小农采用移动农业应用与决策支持系统对农场决策质量的影响。","模拟96户农民数据，描述性分析与多元回归，基于UTAUT框架。","绩效期望、便利条件、社会影响和年龄影响移动应用采用，移动应用与DSS使用正向关联决策质量。","数字乡村与农业信息化","需真实田野数据验证，可探究小农数字工具采用障碍与决策质量提升机制。","openalex","2026-10-07T23:30:38.824339Z",{"total":17,"page":21,"page_size":17,"items":88},[89,172,220,287,368,422],{"id":90,"title":91,"url":92,"summary":93,"summary_zh":94,"content":9,"source_name":95,"source_url":92,"published_at":96,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":97,"score_detail":98,"sources":104,"tags":106,"primary_topic":109,"secondary_topics":114,"classification":120,"search_phrases":152,"slug":155,"view_count":17,"doi":156,"paper":157,"created_at":171},3512,"AI-Driven Agricultural Advisory and Diagnostic Systems for Smallholder Farming: Technical Architectures, Evidence and Deployment Priorities for North-East India","https:\u002F\u002Fdoi.org\u002F10.9734\u002Farja\u002F2026\u002Fv19i4919","Artificial intelligence (AI) is being introduced into agricultural advisory services through machine learning, computer vision, conversational large language models, retrieval-augmented generation and multimodal interfaces. For smallholder farming, the central question is not whether these technologies can produce technically plausible outputs, but whether they can provide locally correct, actionable and safe recommendations under heterogeneous agronomic, linguistic and connectivity conditions. This critical narrative review integrates evidence on digital extension, AI-enabled agricultural advice, image-based diagnosis and responsible digital agriculture, with particular reference to North-East India. Literature published from 1 January 2010 to 17 July 2026 was considered, with emphasis on peer-reviewed field evaluations, technical validation studies, reviews and regionally relevant research. Evidence from digital extension provides the strongest causal baseline: mobile and personalised advisory services frequently improve information recall, agronomic knowledge and adoption of recommended practices, yet effects on yield, profit and welfare are inconsistent. Recent generative-AI studies show that large language models can produce useful agricultural responses, but site-specific rates, timing and local practice remain recurrent failure points. Retrieval grounding and expert feedback improve local relevance, although multi-season farm-level effectiveness evidence remains scarce. Image-based plant-disease systems achieve high accuracy in curated datasets, but performance can deteriorate sharply under field domain shift, class novelty and variable image quality. North-East Indian studies of mobile advisory systems in Meghalaya, Nagaland and Tripura demonstrate a valuable institutional foundation based on interactive voice response, local expert networks and user-centred service design; they do not, however, establish the effectiveness of autonomous AI. The most defensible deployment model is therefore an offline-tolerant, multilingual, multimodal and human-supervised architecture that grounds recommendations in curated regional knowledge, represents uncertainty, preserves provenance and escalates high-risk or out-of-distribution cases. Future research should prioritise prospective district- and season-spanning evaluations that connect model quality to farmer decisions, agronomic outcomes, equity, safety and cost-effectiveness.","人工智能（AI）正通过机器学习、计算机视觉、对话式大语言模型、检索增强生成和多模态界面被引入农业咨询服务。对于小农户而言，核心问题不在于这些技术能否产生技术上看似合理的输出，而在于它们能否在异质的农艺、语言和网络连接条件下提供本地正确、可操作且安全的建议。本批判性叙事综述整合了数字推广、AI赋能的农业建议、基于图像的诊断和负责任数字农业方面的证据，并特别关注印度东北部。本文考察了2010年1月1日至2026年7月17日期间发表的文献，重点关注同行评议的田间评估、技术验证研究、综述及区域相关研究。来自数字推广的证据提供了最强的因果基线：移动化和个性化咨询服务经常改善信息记忆、农艺知识和对推荐措施的采纳，但对产量、利润和福利的影响并不一致。近期生成式AI研究表明，大语言模型能够产生有用的农业回答，但针对具体地点的用量、时机和本地实践仍是反复出现的失败点。检索 grounding 和专家反馈可提高本地相关性，但多季农场层面的有效性证据仍然稀缺。基于图像的植物病害系统在精选数据集上达到高准确率，但在田间域偏移、类别新颖性和图像质量多变的情况下，性能可能急剧下降。印度东北部在梅加拉亚邦、那加兰邦和特里普拉邦开展的移动咨询系统研究展示了基于交互式语音应答、本地专家网络和以用户为中心的服务设计的宝贵制度基础；然而，这些研究并未确立自主AI的有效性。因此，最可辩护的部署模式是一种容忍离线、多语言、多模态且有人工监督的架构，该架构将建议建立在精选的区域知识之上，表征不确定性，保留来源信息，并对高风险或分布外案例进行升级处理。未来研究应优先开展前瞻性的跨区县和跨季节评估，将模型质量与农户决策、农艺结果、公平性、安全性和成本效益联系起来。","Asian Research Journal of Agriculture","2026-09-24T00:00:00Z",80,{"impact":99,"substance":100,"depth":99,"authority":101,"freshness":102,"relevant":21,"comment":103},18,22,13,9,"系统综述AI农业咨询与诊断系统在印度东北小农场景的技术架构与落地证据，指出人机协同、离线多语言与检索增强是可行路径，对智慧农业落地有参考价值。",[105],{"name":95,"url":92},[26,27,107,108,29],"农业人工智能","农业技术推广",{"code":47,"slug":110,"name":111,"path":112,"parent_code":113},"farm-services","社会化服务与规模经营","\u002Ftopic\u002Frural-services\u002Ffarm-services","G",[115],{"code":53,"slug":116,"name":117,"path":118,"parent_code":119},"agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai","I",{"status":121,"locked":14,"version":34,"method":35,"evidence":122},"classified",{"candidates":123,"agricultural_evidence":40},[124,130,139,143,148],{"code":47,"score":125,"supported":40,"title":126,"abstract":128,"tags":129},15,[127],"smallholder",[29,127],[29],{"code":53,"score":13,"supported":40,"title":131,"abstract":132,"tags":138},[],[133,134,135,136,137],"机器学习","计算机视觉","machine learning","artificial intelligence","computer vision",[107],{"code":39,"score":19,"supported":40,"title":140,"abstract":141,"tags":142},[],[43,44],[27],{"code":144,"score":21,"supported":14,"title":145,"abstract":146,"tags":147},"G1",[],[],[108],{"code":59,"score":21,"supported":14,"title":149,"abstract":150,"tags":151},[],[],[26],[153,154],"印度东北部 农业AI 小农户","农业智能诊断 多语言 离线","印度东北部农业AI小农户-3512","10.9734\u002Farja\u002F2026\u002Fv19i4919",{"doi":156,"openalex_id":158,"authors":159,"venue":95,"cited_by_count":78,"oa_url":92,"card":165,"direction":83,"ingested_from":85},"W7214205238",[160,162],{"name":161,"orcid":9},"Pravangkar Boruah",{"name":163,"orcid":164},"Rubul Kumar Bania","https:\u002F\u002Forcid.org\u002F0000-0001-6294-0231",{"tldr":166,"method":167,"finding":168,"direction":169,"opportunity":170},"综述AI农业咨询与诊断系统，聚焦印度东北小农，提出人监督多模态部署架构。","批判性叙述综述，整合2010-2026年数字推广、生成式AI与图像诊断证据。","AI输出技术可行但本地化、安全与田间效果证据不足，需人监督与检索增强。","农业人工智能与决策模型","可开展跨区跨季前瞻评估，连接模型质量与农户决策、产量、公平及成本效益。","2026-09-25T23:30:39.745514Z",{"id":173,"title":174,"url":175,"summary":176,"summary_zh":9,"content":177,"source_name":178,"source_url":9,"published_at":179,"category":180,"cover_url":9,"hotness":13,"is_selected":14,"score":181,"score_detail":182,"sources":187,"tags":189,"primary_topic":9,"secondary_topics":192,"classification":193,"search_phrases":215,"slug":218,"view_count":21,"doi":9,"paper":9,"created_at":219},3374,"数字乡村开启高质量深耕新周期——《行动计划》扭转'重数量轻质量'旧模式","https:\u002F\u002Fwww.rmzxb.com.cn\u002Fc\u002F2026-09-22\u002F3979434.shtml","人民政协报评论指出：历经数年深耕，我国数字乡村建设已完成基础设施全面铺建，但部分地方陷入'建好不用、用而不实'的尴尬：智慧村务平台闲置、涉农系统重复填报、智能设备沦为应付检查的'摆设'。新一轮《行动计划》正式推动数字乡村从'有没有'的普及阶段迈入'好不好'的高质量深耕阶段，跳出'一刀切'固化思维，坚持因地制宜、分类施策、精准赋能。","村村通5G、户户通网络、田间智慧大屏落地……历经数年深耕，我国数字乡村建设已完成基础设施全面铺建。但深入乡村基层不难发现，不少地方的数字化建设陷入“建好不用、用而不实”的尴尬局面：智慧村务平台常年闲置，各类涉农系统重复填报，田间智能设备沦为应付检查的“摆设”。看似遍地开花的数字化场景，始终悬浮于乡土之上，没能真正转化为助农增收、乡村提质的实效。\n\n针对这些基层痛点与发展堵点，《数字乡村高质量发展行动计划（2026—2030年）》正式出台，为未来五年乡村数字化发展锚定全新方向。相较于前期重基建、重覆盖、重样板的粗放式建设，这份新政彻底扭转建设逻辑，推动数字乡村全面告别“重数量、轻质量，重投入、轻运营”的旧模式，正式从“有没有”的普及阶段，迈入“好不好”的高质量深耕阶段，核心要义就是让技术适配乡土、让数据赋能产业、让红利惠及农民。\n\n长期以来，很多人将数字乡村简单等同于直播带货、网络入户，部分地区建设也陷入“城市模板生硬下乡”的误区。不顾乡村产业禀赋差异、城乡发展差距，照搬统一设备、统一系统、统一标准，平原农田、山区果园、水乡养殖千村一面，空心村、薄弱村、富裕村同质化建设。最终导致大量数字化项目“水土不服”，看似光鲜亮眼，却贴合不了农事需求、解决不了基层难题，沦为脱离实际的“面子工程”。\n\n当下数字乡村建设的核心矛盾，早已不是网络不通、设备不足，而是“数据悬浮”与“数字鸿沟升级”。海量的田间生产、乡村治理数据采集后便沉睡后台，部门间信息壁垒森严、数据孤岛突出。基层干部深陷多头填报、重复台账的繁琐工作，数字化反而增添基层负担；广大小农户无法接触、运用数字资源，难以对接市场、规避风险，城乡差距从基础的“网络接入鸿沟”，演变为更深层的“数字应用鸿沟”“素养鸿沟”。\n\n本次行动计划最大的突破，就是跳出“一刀切”的固化建设思维，坚持因地制宜、分类施策、精准赋能。新政明确摒弃同质化建设路径，立足不同乡村的产业特色、资源禀赋、人口结构，定制本土化数字化发展方案。如，粮食主产区聚焦智慧耕种、精准植保、稳产提质；特色村镇深耕农产品溯源、数字电商、乡村文旅数字化；治理薄弱村庄重点推进智慧村务、数字减负、基层提效，真正实现让技术适配农民、贴合乡土，而非让乡村迁就城市标准。\n\n赋能产业提质增收，是本轮数字乡村升级的核心落脚点。新政彻底打破“数字乡村=直播带货”的浅层认知，聚焦培育农业新质生产力，用数字化重构农业全产业链。从春耕土壤监测、无人机智能植保，到田间病虫害预警、生长态势智能管控，再到秋收产量数据分析、农产品全程溯源，以及后端产销精准对接、村级物流网点全覆盖，数字技术贯穿耕、种、管、收、销全链条，推动传统农业彻底告别“看天吃饭、凭经验种地”的粗放模式，迈入“用数据种田、靠精准增效”的智慧新范式。\n\n更具普惠性的是，本轮数字化升级不再是种养大户、新型经营主体的“专属红利”，而是全力向普通小农户倾斜。通过下沉公共数字服务、普及免费数字化工具、开放产销大数据，帮助分散经营的小农户精准对接大市场，有效降低生产成本、规避市场波动风险，让最广泛的农民群体共享数字发展红利，让数字产业成为乡村长效增收的核心引擎。\n\n除了赋能产业发展，数字化升级更聚焦减负基层、惠及民生，重塑乡村治理与公共服务体系。过去，繁杂的线上系统、冗余的台账报表，占用基层干部大量时间精力，让干部深陷线上事务、脱离田间一线。同时，偏远乡村医疗、教育、文化资源薄弱，城乡公共服务不均等问题长期存在。\n\n网线纵横连乡野，数字赋能启新程。未来五年，乡村数字化发展不再比拼硬件数量、样板亮点，而是聚焦实干实效、民生福祉、长效运营。摒弃悬浮化建设、模板化发展的固有弊端，让数据真正扎根阡陌田野，使数字技术深度赋能“三农”发展。\n\n可以说，这场全方位、深层次的数字乡村升级变革，将持续激活乡村沉睡资源、培育农业新质生产力、畅通城乡要素双向流动，为乡村全面振兴、农业强国建设注入持久强劲的数字动能。（记者 李元丽）","人民政协网 2026-09-22","2026-09-22T00:00:00Z","报道",82,{"impact":183,"substance":184,"depth":185,"authority":101,"freshness":19,"relevant":21,"comment":186},26,20,16,"全国性数字乡村新政解读，直指数据悬浮与数字鸿沟痛点，政策信号明确，但属评论性报道、缺乏条款细节。",[188],{"name":178,"url":175},[26,27,190,29,191],"农业新质生产力","智慧村务",[],{"status":33,"locked":14,"version":34,"method":35,"evidence":194},{"candidates":195,"agricultural_evidence":14,"reason":214},[196,201,205,209],{"code":59,"score":197,"supported":40,"title":198,"abstract":199,"tags":200},12,[26],[26],[26],{"code":47,"score":21,"supported":14,"title":202,"abstract":203,"tags":204},[],[],[29],{"code":39,"score":21,"supported":14,"title":206,"abstract":207,"tags":208},[],[],[27],{"code":210,"score":21,"supported":14,"title":211,"abstract":212,"tags":213},"J2",[],[],[190],"农业相关性需要复核",[216,217],"数字乡村 数据孤岛 基层减负","农业新质生产力 数字乡村 智慧农业 智慧村务","数字乡村数据孤岛基层减负-3374","2026-09-25T00:09:26.569518Z",{"id":221,"title":222,"url":223,"summary":224,"summary_zh":225,"content":9,"source_name":226,"source_url":223,"published_at":227,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":228,"score_detail":229,"sources":231,"tags":233,"primary_topic":236,"secondary_topics":237,"classification":238,"search_phrases":259,"slug":262,"view_count":48,"doi":263,"paper":264,"created_at":286},2516,"Scaling climate-smart agriculture through multi-channel advisory services: economic evaluation of weather information systems in sub-Saharan Africa","https:\u002F\u002Fdoi.org\u002F10.1080\u002F14735903.2026.2731746","Smallholder agriculture in sub-Saharan Africa is 95% rain-fed and exposed to climate variability. This post-implementation evaluation applies activity-based costing to weather advisory services delivered by PlantVillage in Kenya and Burkina Faso from 2022 to 2024 across television, radio, short message service (SMS), and community-based channels. A combined investment of USD 551,966 reached 9.41 million farmers at confirmed exposure and an estimated 14.5 million under a moderate-reach scenario (USD 0.038 per farmer). Because extended reach relies on literature-derived multipliers rather than direct measurement, we report a sensitivity range of USD 0.034 to 0.058; even at the conservative end, the cost advantage over Farmer Field Schools (USD 36 to 70 per participant) remains a factor of 620 to 1,200. Mass media achieved the lowest unit cost (USD 0.027 to 0.031), with evidence limited to confirmed exposure, whereas SMS supported intensive engagement at USD 0.60. In a convenience sample of 4,272 registered iShamba users in Kenya, 83.7% (n = 3,577) reported that advisories influenced their decisions, predominantly planting timing (51%) and farm-activity planning (23%). No comparable data were collected in Burkina Faso, and channel contributions appear complementary rather than demonstrably synergistic. Yields and income were not measured; causal impact requires experimental and longitudinal designs.","撒哈拉以南非洲的小农农业95%依赖雨养，易受气候变率影响。本项实施后评估采用作业成本法，对PlantVillage于2022至2024年间在肯尼亚和布基纳法索通过电视、广播、短信服务（SMS）及社区渠道提供的天气咨询服务进行成本核算。合计投资551,966美元，在确认触达口径下覆盖941万农民，在中度触达情景下估计覆盖1,450万农民（每农民0.038美元）。由于扩展触达依赖文献推导的乘数而非直接测量，我们报告0.034至0.058美元的敏感性区间；即使在保守端，相较于农民田间学校（每位参与者36至70美元）的成本优势仍达620至1,200倍。大众媒体的单位成本最低（0.027至0.031美元），但证据仅限于确认触达；短信则以每农民0.60美元支持深度参与。在肯尼亚4,272名注册iShamba用户的便利样本中，83.7%（n = 3,577）报告咨询信息影响了其决策，主要是种植时间（51%）和农事活动规划（23%）。布基纳法索未收集可比数据，各渠道的贡献似乎互补，而非可证明的协同。产量和收入未予测量；因果影响需要实验和纵向设计。","International Journal of Agricultural Sustainability","2026-09-13T00:00:00Z",78,{"impact":99,"substance":100,"depth":99,"authority":101,"freshness":19,"relevant":21,"comment":230},"基于成本核算的天气信息服务大规模评估，单位成本数据与渠道对比结论扎实，对农业气象信息化推广有参考价值。",[232],{"name":226,"url":223},[26,27,234,29,235],"撒哈拉以南非洲","农业气象服务",{"code":47,"slug":110,"name":111,"path":112,"parent_code":113},[],{"status":121,"locked":14,"version":34,"method":35,"evidence":239},{"candidates":240,"agricultural_evidence":40},[241,246,250,254],{"code":39,"score":102,"supported":40,"title":242,"abstract":244,"tags":245},[243],"smart agriculture",[],[27],{"code":47,"score":48,"supported":40,"title":247,"abstract":248,"tags":249},[],[127],[29],{"code":59,"score":21,"supported":14,"title":251,"abstract":252,"tags":253},[],[],[26],{"code":255,"score":21,"supported":14,"title":256,"abstract":257,"tags":258},"J3",[],[],[234],[260,261],"撒哈拉以南非洲 农业气象服务 数字乡村 智慧农业","撒哈拉以南非洲 农业气象服务","撒哈拉以南非洲农业气象服务数字乡村智慧农业-2516","10.1080\u002F14735903.2026.2731746",{"doi":263,"openalex_id":265,"authors":266,"venue":226,"cited_by_count":78,"oa_url":279,"card":280,"direction":285,"ingested_from":85},"W7212902218",[267,270,273,275,277],{"name":268,"orcid":269},"Rimnoma S. Ouedraogo","https:\u002F\u002Forcid.org\u002F0000-0003-3339-300X",{"name":271,"orcid":272},"Bipana Paudel Timilsena","https:\u002F\u002Forcid.org\u002F0000-0003-3584-0563",{"name":274,"orcid":9},"Nebnoma R. Tiendrebeogo",{"name":276,"orcid":9},"Derek Morr",{"name":278,"orcid":9},"D M Hughes","https:\u002F\u002Fwww.tandfonline.com\u002Fdoi\u002Fpdf\u002F10.1080\u002F14735903.2026.2731746?needAccess=true",{"tldr":281,"method":282,"finding":283,"direction":83,"opportunity":284},"评估撒哈拉以南非洲多渠天气信息服务成本与农户决策影响。","采用活动成本法，分析2022-2024年肯尼亚和布基纳法索四类渠道数据。","人均成本仅0.038美元，远低于农民田间学校，83.7%用户称信息影响决策。","需实验与纵向设计验证天气信息对产量收入的因果影响及渠道协同效应。","智慧农业 \u002F 农业物联网","2026-09-15T23:30:10.644934Z",{"id":288,"title":289,"url":290,"summary":291,"summary_zh":292,"content":9,"source_name":293,"source_url":290,"published_at":294,"category":12,"cover_url":9,"hotness":295,"is_selected":14,"score":296,"score_detail":297,"sources":300,"tags":304,"primary_topic":307,"secondary_topics":312,"classification":313,"search_phrases":352,"slug":355,"view_count":68,"doi":356,"paper":357,"created_at":367},2312,"MridAI: Autonomous Edge-to-Conversational IoT for Democratising Precision Agriculture via Neuro-Symbolic Telemetry","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.22724805","While sensor-guided precision agriculture improves water efficiency and curtails chemical run-off, adoption across smallholder farm land in the Global South remains under 1%. Commercial telemetry systems are constrained by high capital acquisition costs (>$300) and complex, dashboard-centric mobile applications that impose heavy cognitive burdens on low-literacy farmers. This paper proposes MridAI, a low-cost (\u003C$45 \u002F ₹3,420 COGS) autonomous in-ground agro-telemetry node coupled with a cloud-based neuro-symbolic artificial intelligence advisory pipeline. The physical layer integrates a multi-parameter Modbus RS485 sensor, an ESP32-C3 micro-controller, an Indian-band 4G LTE Cat-1 modem, and an energy harvesting subsystem within an IP67-rated solar stem. To eliminate measurement errors in high-clay tropical soils (Vertisols), an on-device digital signal-processing pipeline applies an adapted Topp dielectric correction alongside a one-dimensional discrete Kalman filter. To overcome rural telecommunication instability, the firmware implements an asynchronous, non-volatile store-and-forward ring buffer. At the cloud layer, an analytical solver calculates deterministic FAO-56 evapotranspiration deficits and enforces Indian Council of Agricultural Research (ICAR) chemical boundaries, strictly isolating numerical computation from an instruction-tuned Large Language Model (LLM). The generative model functions solely as a linguistic translator, delivering actionable, dialect-adapted recommendations directly via the Meta WhatsApp Cloud API without requiring third-party application downloads. Empirical power-budget modelling confirms indefinite operational autonomy (>240 days without solar irradiance), establishing a scalable paradigm for digital agriculture.","尽管传感器引导的精准农业提高了用水效率并减少了化学品径流，但在全球南方小农农田中的采用率仍不足1%。商业遥测系统受制于高昂的资本购置成本（超过300美元）以及复杂的、以仪表盘为中心的移动应用程序，后者给低识字率农民带来了沉重的认知负担。本文提出MridAI，一种低成本（物料清单成本低于45美元\u002F3，420印度卢比）的自主地下农业遥测节点，并耦合基于云的神经符号人工智能咨询流水线。物理层将多参数Modbus RS485传感器、ESP32-C3微控制器、印度频段4G LTE Cat-1调制解调器以及能量收集子系统集成于IP67防护等级的太阳能杆体内。为消除高黏土热带土壤（变性土）中的测量误差，设备端数字信号处理流水线采用适配的Topp介电校正与一维离散卡尔曼滤波器。为克服农村电信不稳定性，固件实现了异步、非易失性的存储转发环形缓冲区。在云层，分析求解器计算确定性的FAO-56蒸散亏缺，并执行印度农业研究理事会（ICAR）的化学品边界，将数值计算与指令微调的大语言模型（LLM）严格隔离。生成模型仅充当语言翻译器，通过Meta WhatsApp Cloud API直接提供可操作的、适配方言的建议，无需下载第三方应用程序。实证功率预算建模证实了无限期运行自主性（无太阳辐照下超过240天），为数字农业建立了一种可扩展的范式。","Zenodo (CERN European Organization for Nuclear Research)","2026-09-12T00:00:00Z",25,85,{"impact":100,"substance":298,"depth":99,"authority":101,"freshness":102,"relevant":21,"comment":299},23,"面向全球南方小农户的低成本自主土壤遥测节点与神经符号AI咨询管线，方法新颖、成本与能耗数据具体，对农业信息化与普惠数字农业有实质参考价值。",[301,302],{"name":293,"url":290},{"name":293,"url":303},"https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.22724804",[26,27,107,305,29,306],"农业物联网","精准灌溉",{"code":308,"slug":309,"name":310,"path":311,"parent_code":119},"I5","agricultural-iot","物联网传感与边缘计算","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-iot",[],{"status":121,"locked":14,"version":34,"method":35,"evidence":314},{"candidates":315,"agricultural_evidence":40},[316,322,327,332,336,342,348],{"code":39,"score":99,"supported":40,"title":317,"abstract":319,"tags":321},[318],"precision agriculture",[320,43,318,44],"精准农业",[27],{"code":308,"score":102,"supported":40,"title":323,"abstract":325,"tags":326},[324],"IoT",[],[305],{"code":53,"score":19,"supported":40,"title":328,"abstract":329,"tags":331},[],[136,330],"large language model",[107],{"code":47,"score":48,"supported":40,"title":333,"abstract":334,"tags":335},[],[127],[29],{"code":337,"score":48,"supported":40,"title":338,"abstract":339,"tags":341},"H3",[],[340],"evapotranspiration",[306],{"code":343,"score":20,"supported":40,"title":344,"abstract":345,"tags":347},"H2",[],[346],"土壤",[],{"code":59,"score":21,"supported":14,"title":349,"abstract":350,"tags":351},[],[],[26],[353,354],"农业人工智能 农业物联网 数字乡村 智慧农业","农业人工智能 农业物联网","农业人工智能农业物联网数字乡村智慧农业-2312","10.5281\u002Fzenodo.22724805",{"doi":356,"openalex_id":358,"authors":359,"venue":293,"cited_by_count":78,"oa_url":290,"card":362,"direction":285,"ingested_from":85},"W7212357737",[360],{"name":361,"orcid":9},"Pranit Kamble",{"tldr":363,"method":364,"finding":365,"direction":285,"opportunity":366},"提出低成本自主地下物联网节点与神经符号AI咨询管道，让小农户用WhatsApp获取精准农业建议。","ESP32-C3与Modbus传感器、Topp校正和卡尔曼滤波、FAO-56与L","系统成本低于45美元，可独立运行超240天，无需下载应用即可通过WhatsApp获得方言化建议。","可探索低资源语言与方言适配的LLM农业咨询，以及神经符号系统在更多作物和土壤类型中的泛化验证。","2026-09-13T23:30:14.323767Z",{"id":369,"title":370,"url":371,"summary":372,"summary_zh":373,"content":9,"source_name":293,"source_url":371,"published_at":374,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":375,"score_detail":376,"sources":379,"tags":381,"primary_topic":384,"secondary_topics":385,"classification":387,"search_phrases":402,"slug":405,"view_count":78,"doi":406,"paper":407,"created_at":421},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":18,"substance":377,"depth":99,"authority":197,"freshness":13,"relevant":21,"comment":378},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。",[380],{"name":293,"url":371},[27,107,29,382,383],"植物病害识别","可解释性",{"code":47,"slug":110,"name":111,"path":112,"parent_code":113},[386],{"code":53,"slug":116,"name":117,"path":118,"parent_code":119},{"status":121,"locked":14,"version":34,"method":35,"evidence":388},{"candidates":389,"agricultural_evidence":40},[390,394,398],{"code":47,"score":197,"supported":40,"title":391,"abstract":392,"tags":393},[127],[127],[29],{"code":53,"score":13,"supported":40,"title":395,"abstract":396,"tags":397},[137],[],[107,382],{"code":39,"score":21,"supported":14,"title":399,"abstract":400,"tags":401},[],[],[27],[403,404],"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":406,"openalex_id":408,"authors":409,"venue":293,"cited_by_count":78,"oa_url":371,"card":416,"direction":169,"ingested_from":85},"W7223137323",[410,412,414],{"name":411,"orcid":9},"Akshay Ramgoolam",{"name":413,"orcid":9},"Priya Bundhoo Appadoo",{"name":415,"orcid":9},"Jean-Noël LEspérance",{"tldr":417,"method":418,"finding":419,"direction":169,"opportunity":420},"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":423,"title":424,"url":425,"summary":426,"summary_zh":427,"content":9,"source_name":293,"source_url":425,"published_at":374,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":428,"score_detail":429,"sources":431,"tags":433,"primary_topic":436,"secondary_topics":437,"classification":439,"search_phrases":454,"slug":457,"view_count":78,"doi":458,"paper":459,"created_at":473},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":18,"substance":100,"depth":99,"authority":197,"freshness":13,"relevant":21,"comment":430},"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。",[432],{"name":293,"url":425},[27,107,29,434,435],"作物病害识别","模型审计",{"code":47,"slug":110,"name":111,"path":112,"parent_code":113},[438],{"code":53,"slug":116,"name":117,"path":118,"parent_code":119},{"status":121,"locked":14,"version":34,"method":35,"evidence":440},{"candidates":441,"agricultural_evidence":40},[442,446,450],{"code":47,"score":125,"supported":40,"title":443,"abstract":444,"tags":445},[127],[29,127],[29],{"code":53,"score":13,"supported":40,"title":447,"abstract":448,"tags":449},[137],[],[434,107],{"code":39,"score":21,"supported":14,"title":451,"abstract":452,"tags":453},[],[],[27],[455,456],"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":458,"openalex_id":460,"authors":461,"venue":293,"cited_by_count":78,"oa_url":425,"card":468,"direction":169,"ingested_from":85},"W7223088030",[462,464,466],{"name":463,"orcid":9},"Abdi-Basid Idriss Farah",{"name":465,"orcid":9},"Amina Abdillahi Guedi",{"name":467,"orcid":9},"Moussa Omar Youssouf",{"tldr":469,"method":470,"finding":471,"direction":169,"opportunity":472},"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"]