[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-3596":3,"related-3596":70},{"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":8,"secondary_topics":32,"classification":33,"search_phrases":64,"slug":67,"view_count":68,"doi":8,"paper":8,"created_at":69},3596,"中国农科院基因组所商连光团队开发自进化AI科学家'绿洲'助力作物耐盐碱研究——Molecular Plant在线发表","https:\u002F\u002Fagis.caas.cn\u002Fxwzx\u002Fkyjz\u002Fb0960d967c964346b50cb50c4886c209.htm","2026年9月18日，中国农科院基因组所（大鹏湾实验室）商连光团队联合崖州湾国家实验室、盐碱地综合利用国家技术创新中心，在《分子植物（Molecular Plant）》上在线发表题为'OASIS, a self-evolving AI scientist that integrates omics data and literature knowledge for plant stress research'的研究论文。该研究开发了面向植物逆境研究的自进化AI科学家'绿洲'（OASIS）。系统将大语言模型、多智能体协作、植物逆境文献知识库以及组学和遗传数据纳入同一科研流程。'绿洲'将复杂科研任务拆解为多个子任务，由6个AI智能体分工协作完成。研究团队进一步以水稻耐盐为案例，对'绿洲'提出的候选基因开展实验验证，形成了从科学问题到候选假说再到实验验证的完整研究路径。",null,"2026年9月18日，中国农科院基因组所（大鹏湾实验室）商连光团队联合崖州湾国家实验室、盐碱地综合利用国家技术创新中心，在《分子植物（Molecular Plant）》上在线发表题为“OASIS, a self-evolving AI scientist that integrates omics data and literature knowledge for plant stress research”的研究论文。\n\n![Image 1](https:\u002F\u002Fagis.caas.cn\u002Fimages\u002F2026-09\u002Fefa2de3e737741ea8e2ae4ceeed0695f.png)\n\n该研究开发了面向植物逆境研究的自进化AI科学家“绿洲”（OASIS）。系统将大语言模型、多智能体协作、植物逆境文献知识库以及组学和遗传数据纳入同一科研流程，可围绕一个科学问题自主完成任务规划、文献检索、数据查询、文献与组学数据融合推理和结果审查，并能够从历史任务中的失败与纠错中学习可复用经验。研究团队进一步以水稻耐盐为案例，对“绿洲”提出的候选基因开展实验验证，形成了从科学问题到候选假说再到实验验证的完整研究路径。\n\n植物抗逆研究往往不是“缺少信息”，而是证据分散在不同载体中：机制研究主要沉淀在论文里，基因表达、遗传定位、群体变异和种质资源则分布在多个数据库中。研究人员通常需要在大量文献和多类数据之间反复检索、比对和判断，才能把一个科学问题逐步收敛为可供实验验证的假说。通用大语言模型擅长理解和生成文本，但仅依靠模型自身知识，仍难以稳定完成跨数据库、跨数据类型的连续科研分析。\n\n**从“读文献”到“查数据”，让AI进入真实科研工作流**\n\n针对这一问题，“绿洲”将复杂科研任务拆解为多个子任务，由6个AI智能体分工协作完成。系统先识别研究意图并制定任务计划，再根据需要调用文献检索、数据分析和科研数据库等工具，最终由审查与综合模块检查证据覆盖范围和任务完成情况后生成结论。任务规划、文献片段、数据库查询结果和分析记录都会保存在共享工作空间中，便于研究人员回溯和核对。整个过程类似于一个小型科研团队围绕同一科学问题持续协作。\n\n![Image 2](https:\u002F\u002Fagis.caas.cn\u002Fimages\u002F2026-09\u002Ff4ecb718f8d942b89d22b99880d0dbe3.png)\n\n图1 | “绿洲”多智能体协作科研框架。系统围绕科学问题进行任务拆解，并协调文献检索、数据分析、结果审查和综合生成，使不同来源的科研证据能够在同一工作流中被连续调用和融合推理。\n\n支撑这一流程的是专门整理的植物逆境知识和数据资源。“绿洲”目前覆盖拟南芥、水稻、玉米、小麦、大豆和陆地棉6个物种，以及盐、干旱、高温和低温4类主要非生物胁迫。文献知识库收录超过15万篇相关论文摘要和1.5万篇开放获取全文，并持续同步新发表研究；同时接入植物逆境表达数据、GWAS\u002FQTL、群体变异、eQTL和种质资源等结构化信息，使系统不仅能够回答“文献中有哪些报道”，还能够进一步判断“数据是否支持这一推断”。\n\n**OASIS的优势，体现在数据驱动的研究任务中**\n\n为评估系统在真实科研任务中的表现，研究团队构建了面向植物逆境研究的基准测试集PlantStressQA，包含200个贴近真实科研过程的问题，覆盖机制证据、基因ID映射、表达分析、遗传位点分析、品种筛选和群体变异六类任务。\n\n在与多种通用大语言模型的比较中，“绿洲”总体得分为84.0分，参评通用模型得分为33.8至50.1分。在需要实际调用并整合结构化数据的任务中，“绿洲”优势明显，其中群体变异和遗传位点分析两类任务分别比表现最好的通用模型高55.6分和52.3分，在基因ID映射和表达证据分析中也表现出明显优势。\n\n![Image 3](https:\u002F\u002Fagis.caas.cn\u002Fimages\u002F2026-09\u002Fc2b2782477a64a55a3366b44219320d5.png)\n\n图2 | “绿洲”在PlantStressQA上的系统评价。系统在基因ID转换、表达数据查询、遗传位点分析和群体变异分析等数据密集型任务中表现出更明显的优势。\n\n这些结果说明，“绿洲”的核心并不是单纯依赖语言模型“记住更多知识”，而是通过组织科研工具和数据资源，在研究过程中主动查找数据、完成格式转换、比较不同来源的证据，并将这些步骤串联为可追溯的分析过程。\n\n**做错一次之后，下一次还会不会犯同样的错误？**\n\n真实科研分析中，数据库字段不匹配、基因ID格式错误、工具调用顺序不合理等问题并不少见。如果系统每次遇到相似任务都从头试错，不仅效率低，也会造成额外的计算资源消耗。为此，“绿洲”设计了自进化经验学习模块（SEEL），从历史任务中识别“先失败、随后被成功纠正”的执行轨迹，对比错误路径和正确路径，提炼可复用的操作经验，并通过重新执行真实任务检验其有效性。只有验证通过的经验才会进入技能库，供后续相似任务调用。\n\n![Image 4](https:\u002F\u002Fagis.caas.cn\u002Fimages\u002F2026-09\u002F72cd8d054dc14f94b7b9461f4a7f8c46.png)\n\n图3 | SEEL自进化经验学习流程及效果。系统从历史任务的失败与纠错过程中提炼可复用经验，经验证后纳入技能库。\n\n研究团队从PlantStressQA的历史轨迹中筛选出86个包含可复用经验的任务。应用这些经验后，工具调用失败、数据库错误、重复操作和模型处理文本量均有所减少。在一个表达分析任务中，大模型调用次数从75次降至42次，模型处理文本量从112万降至31万，运行时间从476秒缩短到236秒，数据库错误和失败工具调用则从多次降为0次。随着有效经验持续积累，“绿洲”能够在后续相似任务中减少重复试错并优化执行过程。\n\n**最关键的一步：AI提出的候选基因，实验能不能验证？**\n\n基准测试可以衡量AI系统是否“会做题”，但科研价值最终还要回到能否帮助提出值得验证的科学假说。研究团队以水稻耐盐为案例，在未预设具体候选基因的情况下，通过多轮提问引导“绿洲”开展分析。系统先从已有研究中归纳离子稳态与Na⁺\u002FK⁺运输、ABA信号与渗透调节、ROS稳态、MAPK级联和盐感知\u002FCa²⁺信号等主要耐盐调控模块，并整理出284个有文献证据支持的拟南芥调控基因；随后开展跨物种同源映射和功能筛选，再整合水稻耐盐GWAS\u002FQTL等遗传证据，最终逐步确定5个优先候选基因。\n\n![Image 5](https:\u002F\u002Fagis.caas.cn\u002Fimages\u002F2026-09\u002Fa9ba4a0ff48e43bc8239834881fb034c.png)\n\n图4 | “绿洲”辅助筛选水稻耐盐候选基因的多轮对话示例。系统整合耐盐机制文献、水稻同源基因信息和群体遗传证据，逐步筛选出5个优先候选基因。\n\n研究团队根据“绿洲”给出的证据链，选择此前未被直接报道参与水稻耐盐调控的OsPP2a（LOC_Os12g04290）开展实验验证。该基因的拟南芥同源基因TAP46与ABA信号相关，OsPP2a本身位于水稻耐盐meta-QTL区间内，并与一个eQTL信号共定位。两个独立的Ospp2a突变材料在150 mM NaCl处理后均表现出更明显的盐敏感表型，恢复存活率降低，同时Na⁺积累增加、Na⁺\u002FK⁺平衡发生改变。RT-qPCR结果还显示，离子稳态相关基因OsSOS1和OsHKT1;5的表达发生变化，为OsPP2a参与水稻盐胁迫响应提供了遗传和分子层面的证据。\n\n![Image 6](https:\u002F\u002Fagis.caas.cn\u002Fimages\u002F2026-09\u002Fe698f656e5c8488fb80baadceaab1bf0.png)\n\n图5 | “绿洲”优先推荐基因OsPP2a的耐盐功能验证。两个独立突变材料在盐胁迫下表现出更低的恢复存活率和更高的Na⁺\u002FK⁺比值，支持OsPP2a参与水稻耐盐响应。\n\n在这一案例中，研究人员通过多轮提问引导“绿洲”完成机制梳理、跨物种证据迁移、遗传数据查询和候选基因优先级排序，再由实验进一步检验候选假说，形成了“科学问题→证据融合→候选假说→实验验证”相衔接的研究路径。\n\n**从“AI回答科学”到“AI参与科学”**\n\n“绿洲”的目标并不是用AI替代科研人员作出判断，而是把过去需要在多篇文献、多个数据库和不同分析工具之间往返完成的工作，组织成一条可追溯、可复核的科研流程。通过多智能体协作、文献与组学数据融合推理以及自进化经验学习，系统能够帮助研究人员更高效地汇集证据、缩小候选范围并提出更有依据的科学假说。\n\n这项工作为人工智能参与植物抗逆研究提供了一个从“回答问题”走向“参与分析”的实践框架。未来，随着文献、组学数据和科研工具持续扩展，“绿洲”有望进一步服务于作物抗逆机制解析、关键基因挖掘和后续实验研究，为耐盐碱等作物抗逆改良提供智能化研究工具。\n\n基因组所（大鹏湾实验室）商连光研究员和崖州湾国家实验室钱前院士为论文共同通讯作者，基因组所（大鹏湾实验室）博士后韩漾和副研究员魏华为论文共同第一作者。该研究得到国家重点研发计划和国家自然科学基金的资助。\n\n平台入口：[https:\u002F\u002Fwww.oasis.ac.cn](https:\u002F\u002Fwww.oasis.ac.cn\u002F)\n\n原文链接：[https:\u002F\u002Fwww.cell.com\u002Fmolecular-plant\u002Fabstract\u002FS1674-2052(26)00305-9](https:\u002F\u002Fwww.cell.com\u002Fmolecular-plant\u002Fabstract\u002FS1674-2052(26)00305-9)","中国农科院基因组研究所 2026-09-18","2026-09-18T00:00:00Z","报道",10,true,90,{"impact":17,"substance":18,"depth":19,"authority":20,"freshness":21,"relevant":22,"comment":23},26,24,19,15,6,1,"国家级科研机构在《分子植物》发表的自进化AI科学家系统，实现文献与组学数据融合推理并经水稻耐盐基因实验验证，属农业人工智能与智慧育种领域的标志性突破。",[25],{"name":10,"url":6},[27,28,29,30,31],"农业人工智能","水稻","多智能体","耐盐碱","智慧育种",[],{"status":34,"locked":35,"version":36,"method":37,"evidence":38},"needs_review",false,"2026-10-22.2","rules",{"candidates":39,"agricultural_evidence":14,"reason":63},[40,46,52,58],{"code":41,"score":42,"supported":14,"title":43,"abstract":44,"tags":45},"A1",4,[],[28],[28],{"code":47,"score":42,"supported":14,"title":48,"abstract":49,"tags":51},"H2",[],[50],"盐碱地",[30],{"code":53,"score":54,"supported":35,"title":55,"abstract":56,"tags":57},"I2",2,[],[],[27,29],{"code":59,"score":22,"supported":35,"title":60,"abstract":61,"tags":62},"F2",[],[],[31],"多个产业证据相当，需要确认主归属",[65,66],"中国农科院基因组所 绿洲 OASIS","商连光 水稻耐盐 OsPP2a","中国农科院基因组所绿洲OASIS-3596",7,"2026-09-27T00:05:16.998310Z",{"total":21,"page":22,"page_size":21,"items":71},[72,171,252,326,409,463],{"id":73,"title":74,"url":75,"summary":76,"summary_zh":77,"content":8,"source_name":78,"source_url":75,"published_at":79,"category":80,"cover_url":8,"hotness":13,"is_selected":35,"score":81,"score_detail":82,"sources":88,"tags":90,"primary_topic":94,"secondary_topics":99,"classification":106,"search_phrases":146,"slug":149,"view_count":150,"doi":151,"paper":152,"created_at":170},5496,"Multi-Crop Irrigation Need Classification Based on Machine Learning with an Indonesian-Language GenAI-UX Recommendation Layer","https:\u002F\u002Fdoi.org\u002F10.30871\u002Fjaic.v10i5.13774","Inefficient irrigation water use remains a critical challenge in Indonesian agriculture, and the complexity of raw sensor data hinders adoption by smallholder farmers who lack data-analysis skills. This study develops an intelligent irrigation Decision Support System that integrates interpretable Machine Learning with an Indonesian-language GenAI-UX recommendation layer, using a publicly available multi-crop irrigation scheduling dataset (501 records, nine crops) with an analytical focus on paddy and maize. Four components are combined into one pipeline: (i) irrigation-need classification compared across six algorithms; (ii) regression of a continuous water-requirement score; (iii) sensor-anomaly detection using Isolation Forest cross-validated with the Interquartile Range method; and (iv) a rule-based natural-language recommendation layer whose operating point is optimized by a Genetic Algorithm. Validity was safeguarded through stratified five-fold cross-validation and leakage-controlled preprocessing. Gradient Boosting achieved the best classification performance (F1-score 0.980; cross-validated F1 0.963), Random Forest Regressor reached R square of 0.807, Isolation Forest agreed with IQR outliers at 95.2%, and the Genetic Algorithm attained a confidence of 0.993 within realistic agronomic bounds. In an illustrative usability evaluation, the GenAI-UX interface obtained the highest System Usability Scale score (Good category). The results indicate that pairing accurate, interpretable models with a language-based delivery layer can transform complex sensor data into actionable irrigation recommendations, lowering the adoption barrier for non-technical users. Field validation with real respondents and edge deployment are identified as the next steps.","灌溉用水效率低下仍是印度尼西亚农业面临的一项关键挑战，而原始传感器数据的复杂性阻碍了缺乏数据分析技能的小农户采用相关技术。本研究开发了一种智能灌溉决策支持系统（Decision Support System），将可解释的机器学习与印度尼西亚语生成式人工智能用户体验（GenAI-UX）推荐层相结合，使用公开可用的多作物灌溉调度数据集（501条记录，九种作物），并以水稻和玉米为分析重点。四个组件被整合为一条流水线：（i）在六种算法之间比较灌溉需求分类；（ii）对连续需水量得分进行回归；（iii）使用孤立森林（Isolation Forest）进行传感器异常检测，并与四分位距（IQR）方法交叉验证；（iv）基于规则的自然语言推荐层，其操作点由遗传算法优化。通过分层五折交叉验证和防泄漏预处理保障了有效性。梯度提升（Gradient Boosting）取得了最佳分类性能（F1分数0.980；交叉验证F1为0.963），随机森林回归器（Random Forest Regressor）达到R平方0.807，孤立森林与IQR异常值的一致率为95.2%，遗传算法在现实农艺范围内达到0.993的置信度。在一项示例性可用性评估中，GenAI-UX界面获得了最高的系统可用性量表（System Usability Scale）得分（良好类别）。结果表明，将准确、可解释的模型与基于语言的交付层相结合，能够将复杂的传感器数据转化为可操作的灌溉建议，从而降低非技术用户的采用门槛。与真实受访者进行田间验证以及边缘部署被确定为下一步工作。","Journal of Applied Informatics and Computing","2026-10-19T00:00:00Z","论文",69,{"impact":83,"substance":84,"depth":85,"authority":83,"freshness":86,"relevant":22,"comment":87},12,20,17,8,"方法组合完整、指标扎实的农业AI决策支持论文，但属印尼小农户场景的实验室级研究，公共影响有限，适合作为技术参考而非每日精选头条。",[89],{"name":78,"url":75},[91,27,28,92,93],"智慧农业","玉米","灌溉决策",{"code":53,"slug":95,"name":96,"path":97,"parent_code":98},"agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai","I",[100],{"code":101,"slug":102,"name":103,"path":104,"parent_code":105},"H3","irrigation","水资源与灌溉","\u002Ftopic\u002Fenvironment\u002Firrigation","H",{"status":107,"locked":35,"version":36,"method":37,"evidence":108},"classified",{"candidates":109,"agricultural_evidence":14},[110,118,123,128,134,141],{"code":53,"score":111,"supported":14,"title":112,"abstract":114,"tags":117},18,[113],"machine learning",[115,113,116],"机器学习","decision support",[27],{"code":101,"score":20,"supported":14,"title":119,"abstract":120,"tags":122},[102],[121,102],"灌溉",[93],{"code":41,"score":68,"supported":14,"title":124,"abstract":125,"tags":127},[],[28,126],"paddy",[28],{"code":129,"score":68,"supported":14,"title":130,"abstract":131,"tags":133},"A3",[],[92,132],"maize",[92],{"code":135,"score":21,"supported":14,"title":136,"abstract":137,"tags":140},"G2",[],[138,139],"小农户","smallholder",[],{"code":142,"score":22,"supported":35,"title":143,"abstract":144,"tags":145},"I1",[],[],[91],[147,148],"印尼 多作物 灌溉决策","GenAI 灌溉推荐","印尼多作物灌溉决策-5496",0,"10.30871\u002Fjaic.v10i5.13774",{"doi":151,"openalex_id":153,"authors":154,"venue":78,"cited_by_count":150,"oa_url":75,"card":162,"direction":168,"ingested_from":169},"W7223016583",[155,158,160],{"name":156,"orcid":157},"Muhammad Rusdi","https:\u002F\u002Forcid.org\u002F0000-0002-1048-5000",{"name":159,"orcid":8},"Zaenuddin Zaenuddin",{"name":161,"orcid":8},"Hendra Sanjaya",{"tldr":163,"method":164,"finding":165,"direction":166,"opportunity":167},"构建融合可解释机器学习与印尼语GenAI推荐层的智能灌溉决策支持系统。","六种分类算法、回归、孤立森林异常检测与遗传算法优化规则推荐层。","梯度提升分类F1达0.980，GenAI界面可用性最佳，降低非技术用户使用门槛。","智慧农业 \u002F 农业物联网","可探索多语言本地化GenAI推荐层与边缘部署，并开展真实农户田间验证。","农业人工智能与决策模型","openalex","2026-10-21T23:30:46.441864Z",{"id":172,"title":173,"url":174,"summary":175,"summary_zh":176,"content":8,"source_name":177,"source_url":174,"published_at":178,"category":80,"cover_url":8,"hotness":13,"is_selected":35,"score":179,"score_detail":180,"sources":183,"tags":185,"primary_topic":188,"secondary_topics":192,"classification":198,"search_phrases":232,"slug":235,"view_count":150,"doi":236,"paper":237,"created_at":251},5394,"Rice Yield Prediction Using Multi-Modal Regression Learning with Optimized Ensemble Models","https:\u002F\u002Fdoi.org\u002F10.3390\u002Feng7100572","Accurate rice yield estimation is important for agricultural planning, resource allocation, and food security assessment, but yield depends on interacting climatic, vegetation, remote sensing, seasonal, and crop growth factors. This study presents a Multi-Modal Regression Learning (MMRL) framework for continuous rice yield prediction. The framework integrates rainfall, temperature, humidity, normalized difference vegetation index (NDVI), VH synthetic aperture radar (SAR) backscatter, season, rice cultivation area, and leaf wetness information. Predictor relevance is assessed before model development, and multiple regression learners are trained using a common feature set. Their validation predictions are combined through non-negative ensemble weights constrained to sum to one, thereby producing a single optimized prediction. Experiments were performed on a modeled spatio-temporal agricultural dataset representing Karnataka conditions, with 10,000 observations used for the regression comparison. The proposed MMRL ensemble achieved an R2 of 0.9515, an RMSE of 0.6334 tons\u002Fha, an MAE of 0.5075 tons\u002Fha, and an MAPE of 9.81%, outperforming the evaluated individual regression models. Diagnostic plots further indicated close agreement between predicted and reference yields across most of the evaluated range. The MMRL workflow was also evaluated on a separately reconstructed synthetic dataset generated with fully specified distributions, sampling rules, noise terms, a yield generation equation, and fixed random seeds.","准确的水稻产量估算对农业规划、资源配置和粮食安全评估具有重要意义，但产量取决于气候、植被、遥感、季节和作物生长等多种因素的相互作用。本研究提出了一种用于连续水稻产量预测的多模态回归学习（Multi-Modal Regression Learning，MMRL）框架。该框架整合了降雨量、温度、湿度、归一化植被指数（NDVI）、VH合成孔径雷达（SAR）后向散射、季节、水稻种植面积和叶片湿润信息。在模型开发之前评估预测变量的相关性，并使用共同特征集训练多个回归学习器。其验证预测结果通过约束为和等于1的非负集成权重进行组合，从而生成单一优化预测。实验在代表卡纳塔克邦条件的建模时空农业数据集上进行，使用10,000个观测值进行回归比较。所提出的MMRL集成模型实现了R²为0.9515、RMSE为0.6334吨\u002F公顷、MAE为0.5075吨\u002F公顷、MAPE为9.81%，优于所评估的各个回归模型。诊断图进一步表明，在大多数评估范围内，预测产量与参考产量高度一致。MMRL工作流程还在一个单独重建的合成数据集上进行了评估，该数据集使用完全指定的分布、采样规则、噪声项、产量生成方程和固定随机种子生成。","Eng—Advances in Engineering","2026-10-18T00:00:00Z",78,{"impact":111,"substance":181,"depth":111,"authority":83,"freshness":86,"relevant":22,"comment":182},22,"多模态回归集成方法在水稻估产上取得较高精度，方法新颖、数据规模明确，对农业信息化有参考价值，但属学术论文且数据集为模拟数据，产业影响有限。",[184],{"name":177,"url":174},[91,27,28,186,187],"产量预测","遥感",{"code":41,"slug":189,"name":28,"path":190,"parent_code":191},"rice","\u002Ftopic\u002Ffield-crops\u002Frice","A",[193],{"code":194,"slug":195,"name":196,"path":197,"parent_code":98},"I3","remote-sensing","遥感与农情监测","\u002Ftopic\u002Fdigital-agriculture\u002Fremote-sensing",{"status":107,"locked":35,"version":36,"method":37,"evidence":199},{"candidates":200,"agricultural_evidence":14},[201,205,210,217,224,228],{"code":41,"score":20,"supported":14,"title":202,"abstract":203,"tags":204},[189],[28,189],[28],{"code":194,"score":86,"supported":14,"title":206,"abstract":207,"tags":209},[],[187,208],"remote sensing",[186,187],{"code":211,"score":21,"supported":14,"title":212,"abstract":213,"tags":216},"A8",[],[214,215],"粮食安全","food security",[],{"code":218,"score":219,"supported":14,"title":220,"abstract":221,"tags":223},"J1",3,[],[222],"农业规划",[],{"code":142,"score":22,"supported":35,"title":225,"abstract":226,"tags":227},[],[],[91],{"code":53,"score":22,"supported":35,"title":229,"abstract":230,"tags":231},[],[],[27],[233,234],"水稻 产量预测 多模态","NDVI SAR 水稻估产","水稻产量预测多模态-5394","10.3390\u002Feng7100572",{"doi":236,"openalex_id":238,"authors":239,"venue":177,"cited_by_count":150,"oa_url":174,"card":245,"direction":249,"ingested_from":169},"W7222942064",[240,242],{"name":241,"orcid":8},"Nayana Venkatesh",{"name":243,"orcid":244},"Evangelin Geetha Devaraj","https:\u002F\u002Forcid.org\u002F0000-0002-7253-4314",{"tldr":246,"method":247,"finding":248,"direction":249,"opportunity":250},"构建多模态回归学习框架，融合气候、遥感与作物因子，实现水稻产量连续预测。","融合降雨、温度、NDVI、SAR后向散射等多源特征，用非负权重集成多个回归器。","集成模型R²达0.9515，RMSE为0.6334吨\u002F公顷，优于各单一回归模型。","农业遥感与作物表型","可引入真实田间多时相遥感与深度时序模型，验证合成数据结论在真实产区的泛化性。","2026-10-20T23:30:19.292634Z",{"id":253,"title":254,"url":255,"summary":256,"summary_zh":257,"content":8,"source_name":258,"source_url":255,"published_at":79,"category":80,"cover_url":8,"hotness":13,"is_selected":35,"score":259,"score_detail":260,"sources":264,"tags":266,"primary_topic":269,"secondary_topics":270,"classification":271,"search_phrases":308,"slug":311,"view_count":150,"doi":312,"paper":313,"created_at":325},5388,"A digital twin framework for proactive disease management in paddy cultivation using spatio-temporal graph neural networks","https:\u002F\u002Fdoi.org\u002F10.1080\u002F20421338.2026.2707951","Rice production is highly affected by disease outbreaks, particularly in Asian agricultural systems, resulting in major yield losses and excessive pesticide usage. This study proposes a digital twin framework for paddy cultivation that integrates predictive disease modelling with real-time crop phenology simulation to support proactive disease management. The framework utilizes historical phenological records, IoT-based microclimatic sensor data, irrigation information, and field management records collected from 150 hectares of paddy fields in Maharashtra, India. A Spatio-Temporal Graph Neural Network (ST-GNN), integrating Graph Convolutional Networks (GCNs) and Long Short-Term Memory (LSTM) networks, is employed to model spatial disease propagation and temporal crop dynamics for predicting blast, brown planthopper, and bacterial leaf blight outbreaks up to 14 days in advance. Experimental results demonstrate 87.3% prediction accuracy, 23.7% reduction in pesticide usage, and 31.2% reduction in crop losses, while outperforming traditional Random Forest and SVM models in F1-score performance. The proposed framework enables real-time decision support for sustainable precision agriculture and optimized resource management.","水稻生产极易受到病害暴发的影响，在亚洲农业体系中尤为突出，造成重大产量损失和农药过度使用。本研究提出一个水稻栽培数字孪生（digital twin）框架，将预测性病害建模与实时作物物候模拟相结合，以支持主动式病害管理。该框架利用从印度马哈拉施特拉邦150公顷稻田采集的历史物候记录、基于物联网的微气候传感器数据、灌溉信息及田间管理记录。采用融合图卷积网络（Graph Convolutional Networks, GCN）与长短期记忆网络（Long Short-Term Memory, LSTM）的时空图神经网络（Spatio-Temporal Graph Neural Network, ST-GNN），对病害空间传播与作物时序动态进行建模，以提前14天预测稻瘟病、褐飞虱和白叶枯病的暴发。实验结果表明，预测准确率达87.3%，农药使用量减少23.7%，作物损失减少31.2%，且在F1分数上优于传统随机森林（Random Forest）和支持向量机（SVM）模型。所提框架可为可持续精准农业和优化资源管理提供实时决策支持。","African Journal of Science Technology Innovation and Development",80,{"impact":111,"substance":181,"depth":111,"authority":261,"freshness":262,"relevant":22,"comment":263},13,9,"方法新颖、数据规模明确且量化效果显著，属智慧农业细分领域实质进展，值得进入每日精选。",[265],{"name":258,"url":255},[91,27,28,267,268],"数字孪生","病虫害预警",{"code":41,"slug":189,"name":28,"path":190,"parent_code":191},[],{"status":107,"locked":35,"version":36,"method":37,"evidence":272},{"candidates":273,"agricultural_evidence":14},[274,281,286,293,297,304],{"code":142,"score":19,"supported":14,"title":275,"abstract":277,"tags":280},[276],"digital twin",[278,267,279,276],"精准农业","precision agriculture",[267,91],{"code":41,"score":111,"supported":14,"title":282,"abstract":283,"tags":285},[126],[28,284,189,126],"稻田",[28],{"code":287,"score":21,"supported":14,"title":288,"abstract":289,"tags":292},"F4",[],[290,291],"农药","pesticide",[],{"code":101,"score":21,"supported":14,"title":294,"abstract":295,"tags":296},[],[121,102],[],{"code":298,"score":21,"supported":14,"title":299,"abstract":300,"tags":303},"I5",[],[301,302],"物联网","IoT",[],{"code":53,"score":42,"supported":14,"title":305,"abstract":306,"tags":307},[],[116],[27],[309,310],"水稻 病害预测 数字孪生","ST-GNN 水稻 病虫害","水稻病害预测数字孪生-5388","10.1080\u002F20421338.2026.2707951",{"doi":312,"openalex_id":314,"authors":315,"venue":258,"cited_by_count":150,"oa_url":8,"card":320,"direction":166,"ingested_from":169},"W7222969403",[316,318],{"name":317,"orcid":8},"Jyoti Rani",{"name":319,"orcid":8},"Dr.  Narendra S.  Jadhav",{"tldr":321,"method":322,"finding":323,"direction":168,"opportunity":324},"构建水稻数字孪生框架，用时空图神经网络提前14天预测病害。","融合物联网微气候、物候与管理数据，用GCN+LSTM的ST-GNN建模。","预测准确率87.3%，农药减23.7%、损失减31.2%，优于RF与SVM。","可探索多尺度时空图建模与数字孪生闭环决策，并迁移至其他作物与区域。","2026-10-20T23:30:12.636299Z",{"id":327,"title":328,"url":329,"summary":330,"summary_zh":331,"content":8,"source_name":332,"source_url":329,"published_at":333,"category":80,"cover_url":8,"hotness":13,"is_selected":35,"score":334,"score_detail":335,"sources":337,"tags":339,"primary_topic":343,"secondary_topics":344,"classification":345,"search_phrases":384,"slug":387,"view_count":150,"doi":388,"paper":389,"created_at":408},5161,"RECENT ADVANCES IN DIRECT SEEDED RICE: PHYSIOLOGICAL ADAPTATION, CROP MANAGEMENT, AND GENETIC IMPROVEMENT","https:\u002F\u002Fdoi.org\u002F10.4238\u002Fvknnq971","Owing to insufficient water, labour shortages and accelerating climate change, direct seeded rice (DSR) has emerged as a feasible and resource-efficient alternative to conventional transplanted puddled rice (TPR) systems, conserving water and reducing methane emissions. However, the initial growth phases are frequently impeded by inadequate seedling emergence, intensive weed competition, nutritional deficiencies and exposure to variable soil moisture and aeration conditions in some DSR systems, including low nitrogen use efficiency, which can impair the establishment of a healthy crop stand and stability of its yield under DSR. Identifying and phenotyping rice germplasm for critical physiological traits is essential for developing cultivars that thrive in DSR environments. The physiological basis of genotypes adaption to DSR conditions need to be integrated with phenotyping and breeding strategies. This review synthesizes the knowledge of the physiological traits that facilitate DSR adaptation evaluates both conventional and high-throughput phenotyping (HTP) techniques — digital imaging, canopy temperature measurement, chlorophyll fluorescence analysis, and remote sensing technologies to precisely evaluate these traits in field and controlled environment and examines how quantitative trait loci (QTL) mapping, genome-wide association studies (GWAS), and marker-assisted selection (MAS), significantly facilitates the development of rice varieties suitable for DSR. The review highlights the importance of multi-trait phenotyping frameworks and multi-environment testing in accelerating genetic improvement and ensuring sustainable rice production in direct seeded agricultural systems.","由于水资源不足、劳动力短缺以及气候变化加剧，直播稻（direct seeded rice, DSR）已成为传统移栽水田稻（transplanted puddled rice, TPR）体系的一种可行且资源高效的替代方案，能够节约用水并减少甲烷排放。然而，在一些直播稻体系中，初期生长阶段常常受到出苗不齐、杂草竞争激烈、营养缺乏以及土壤水分和通气条件多变等因素的制约，包括氮利用效率低，这些问题可能损害健康作物群体的建成及其在直播稻条件下产量的稳定性。鉴定和表型分析水稻种质的关键生理性状，对于培育在直播稻环境中生长良好的品种至关重要。基因型适应直播稻条件的生理基础需要与表型分析和育种策略相结合。本综述综合了促进直播稻适应的生理性状相关知识，评估了传统和高通量表型分析（high-throughput phenotyping, HTP）技术——数字成像、冠层温度测量、叶绿素荧光分析和遥感技术——在田间和受控环境中精确评价这些性状的作用，并探讨了数量性状位点（quantitative trait loci, QTL）定位、全基因组关联分析（genome-wide association studies, GWAS）和标记辅助选择（marker-assisted selection, MAS）如何显著促进适宜直播稻的水稻品种选育。本综述强调了多性状表型分析框架和多环境测试在加速遗传改良、确保直播稻农业生产体系可持续水稻生产方面的重要性。","Genetics and Molecular Research","2026-10-16T00:00:00Z",77,{"impact":111,"substance":84,"depth":85,"authority":261,"freshness":262,"relevant":22,"comment":336},"系统综述直播稻生理适应、表型技术与分子育种进展，对水稻轻简化栽培与智慧育种有参考价值，但属综述类论文，影响偏行业细分。",[338],{"name":332,"url":329},[28,340,31,341,342],"种业振兴","表型组学","直播稻",{"code":41,"slug":189,"name":28,"path":190,"parent_code":191},[],{"status":107,"locked":35,"version":36,"method":37,"evidence":346},{"candidates":347,"agricultural_evidence":14},[348,352,358,365,369,375,380],{"code":41,"score":20,"supported":14,"title":349,"abstract":350,"tags":351},[189],[28,189],[28],{"code":47,"score":21,"supported":14,"title":353,"abstract":354,"tags":357},[],[355,356],"土壤","soil",[],{"code":359,"score":21,"supported":14,"title":360,"abstract":361,"tags":364},"H7",[],[362,363],"气候变化","climate change",[],{"code":194,"score":21,"supported":14,"title":366,"abstract":367,"tags":368},[],[187,208],[],{"code":370,"score":42,"supported":14,"title":371,"abstract":372,"tags":374},"F1",[],[373],"germplasm",[340],{"code":101,"score":219,"supported":14,"title":376,"abstract":377,"tags":379},[],[378],"水资源",[],{"code":59,"score":22,"supported":35,"title":381,"abstract":382,"tags":383},[],[],[31],[385,386],"直播稻 生理适应 遗传改良","水稻 高通量表型 分子标记","直播稻生理适应遗传改良-5161","10.4238\u002Fvknnq971",{"doi":388,"openalex_id":390,"authors":391,"venue":332,"cited_by_count":150,"oa_url":329,"card":403,"direction":249,"ingested_from":169},"W7222406581",[392,394,397,399,401],{"name":393,"orcid":8},"Vaishnavi. T S",{"name":395,"orcid":396},"N. Sritharan","https:\u002F\u002Forcid.org\u002F0000-0002-4016-7281",{"name":398,"orcid":8},"V Ravichandran",{"name":400,"orcid":8},"Suresh. R",{"name":402,"orcid":8},"Kannan. P",{"tldr":404,"method":405,"finding":406,"direction":249,"opportunity":407},"综述直播稻生理适应、栽培管理与遗传改良，强调表型组与育种结合。","整合传统与高通量表型技术及QTL\u002FGWAS\u002FMAS育种策略。","多性状表型框架和多环境测试可加速直播稻遗传改良。","可开发面向直播稻苗期抗逆与氮效的高通量表型平台并耦合基因组选择。","2026-10-17T23:30:20.384086Z",{"id":410,"title":411,"url":412,"summary":413,"summary_zh":414,"content":8,"source_name":415,"source_url":412,"published_at":416,"category":80,"cover_url":8,"hotness":13,"is_selected":35,"score":259,"score_detail":417,"sources":419,"tags":421,"primary_topic":424,"secondary_topics":425,"classification":427,"search_phrases":442,"slug":445,"view_count":150,"doi":446,"paper":447,"created_at":462},4802,"A Privacy-Preserving Federated Learning Framework for Paddy Leaf Disease Detection Using an Enhanced Pretrained Novel EfficientNet-Lite4 Model","https:\u002F\u002Fdoi.org\u002F10.15832\u002Fankutbd.1837705","The correct and prompt detection of leaf diseases in paddy crops remains essential for protecting agricultural output and yield stability and reducing major losses that occur during severe disease outbreaks. The widespread use of conventional centralized machine learning techniques becomes impractical for real-world agricultural applications, as their effectiveness diminishes under three main conditions: data privacy limitations, delayed transmission of information, along with reduced performance accuracy from non-independent and identically distributed (non-IID) data properties across different agro-climatic regions. This research develops a decentralized diagnostic system which combines Federated Learning with an Enhanced Pretrained Novel EfficientNet-Lite4 model designed for paddy leaf disease classification. The Paddy Doctor dataset served as training material for the system which contained 16,225 RGB images that classified 12 disease types alongside healthy categories. The proposed model reached 98.7% classification accuracy and 98.5% precision and 98.6% F1-score while maintaining a final loss of 0.21 after 100 communication rounds. The system underwent field testing which demonstrated its ability to generate accurate diagnostic results in real time in various working conditions. The study demonstrates how privacy preserving AI-based paddy disease management works in practice and recommends local agricultural institutions to build decentralized learning systems. The suggested system will give farmers direct access to field level decision support and facilitate the creation of sustainable digital pest management systems required to ensure food supply.","正确且及时地检测水稻作物叶片病害，对于保护农业产出和产量稳定、减少病害严重暴发期间发生的重大损失仍然至关重要。传统集中式机器学习技术的广泛使用在现实农业应用中变得不切实际，因为在三种主要条件下其有效性会下降：数据隐私限制、信息传输延迟，以及不同农业气候区域间非独立同分布（non-IID）数据特性导致的性能准确率降低。本研究开发了一种去中心化诊断系统，将联邦学习与一种增强的预训练新型高效网络模型EfficientNet-Lite4相结合，用于水稻叶片病害分类。Paddy Doctor数据集作为该系统的训练材料，包含16,225张RGB图像，涵盖12种病害类型以及健康类别。所提出的模型在100轮通信后达到了98.7%的分类准确率、98.5%的精确率和98.6%的F1分数，同时最终损失保持在0.21。该系统经过实地测试，证明其能够在各种工作条件下实时生成准确的诊断结果。该研究展示了基于隐私保护人工智能的水稻病害管理如何在实践中运作，并建议地方农业机构构建去中心化学习系统。所建议的系统将为农民提供直接的田间决策支持，并促进建立确保粮食供应所需的可持续数字病虫害管理系统。","Tarım Bilimleri Dergisi","2026-10-12T00:00:00Z",{"impact":111,"substance":181,"depth":111,"authority":261,"freshness":262,"relevant":22,"comment":418},"将联邦学习与轻量级EfficientNet-Lite4结合用于水稻叶病识别，兼顾隐私保护与98.7%高精度，对智慧农业落地有参考价值。",[420],{"name":415,"url":412},[91,27,28,422,423],"病虫害识别","联邦学习",{"code":41,"slug":189,"name":28,"path":190,"parent_code":191},[426],{"code":53,"slug":95,"name":96,"path":97,"parent_code":98},{"status":107,"locked":35,"version":36,"method":37,"evidence":428},{"candidates":429,"agricultural_evidence":14},[430,434,438],{"code":41,"score":20,"supported":14,"title":431,"abstract":432,"tags":433},[126],[28,126],[28],{"code":53,"score":83,"supported":14,"title":435,"abstract":436,"tags":437},[],[115,113,116],[27,422,423],{"code":142,"score":22,"supported":35,"title":439,"abstract":440,"tags":441},[],[],[91],[443,444],"Paddy Doctor 数据集 水稻病害","EfficientNet-Lite4 水稻叶片病害","PaddyDoctor数据集水稻病害-4802","10.15832\u002Fankutbd.1837705",{"doi":446,"openalex_id":448,"authors":449,"venue":415,"cited_by_count":150,"oa_url":456,"card":457,"direction":166,"ingested_from":169},"W7221522859",[450,453],{"name":451,"orcid":452},"B Johnson","https:\u002F\u002Forcid.org\u002F0009-0009-8281-4051",{"name":454,"orcid":455},"Chandrakumar Thangavel","https:\u002F\u002Forcid.org\u002F0000-0002-1186-5988","https:\u002F\u002Fdergipark.org.tr\u002Fen\u002Fdownload\u002Farticle-file\u002F5486058",{"tldr":458,"method":459,"finding":460,"direction":168,"opportunity":461},"提出联邦学习结合改进EfficientNet-Lite4的水稻叶片病害检测框架，兼顾隐私与精度。","联邦学习+预训练EfficientNet-Lite4，用Paddy Doctor","12类病害分类准确率达98.7%，100轮通信后损失0.21，支持实时诊断。","可探索非IID数据下联邦聚合优化与轻量化模型在边缘设备的部署研究。","2026-10-13T23:30:08.809781Z",{"id":464,"title":465,"url":466,"summary":467,"summary_zh":8,"content":8,"source_name":468,"source_url":466,"published_at":469,"category":80,"cover_url":8,"hotness":13,"is_selected":35,"score":470,"score_detail":471,"sources":473,"tags":475,"primary_topic":477,"secondary_topics":478,"classification":481,"search_phrases":501,"slug":504,"view_count":22,"doi":505,"paper":506,"created_at":538},4500,"Integrating UAV-Based Remote Sensing and Deep Learning for Early Detection of Bacterial Leaf Blight in Rice","https:\u002F\u002Fdoi.org\u002F10.21203\u002Frs.3.rs-11163942\u002Fv1","Integrating UAV-Based Remote Sensing and Deep Learning for Early Detection of Bacterial Leaf Blight in Rice。Research Square","Research Square","2026-10-08T00:00:00Z",74,{"impact":111,"substance":84,"depth":111,"authority":13,"freshness":86,"relevant":22,"comment":472},"将无人机遥感与深度学习结合用于水稻白叶枯病早期检测，方法有创新、结论具应用价值，但属预印本且为细分技术进展，适合进入主题聚合而非头条精选。",[474],{"name":468,"url":466},[91,27,28,187,476],"病害检测",{"code":41,"slug":189,"name":28,"path":190,"parent_code":191},[479,480],{"code":53,"slug":95,"name":96,"path":97,"parent_code":98},{"code":194,"slug":195,"name":196,"path":197,"parent_code":98},{"status":107,"locked":35,"version":36,"method":37,"evidence":482},{"candidates":483,"agricultural_evidence":14},[484,488,493,497],{"code":41,"score":83,"supported":14,"title":485,"abstract":486,"tags":487},[189],[189],[28],{"code":53,"score":83,"supported":14,"title":489,"abstract":491,"tags":492},[490],"deep learning",[490],[27],{"code":194,"score":83,"supported":14,"title":494,"abstract":495,"tags":496},[208],[208],[187],{"code":142,"score":22,"supported":35,"title":498,"abstract":499,"tags":500},[],[],[91],[502,503],"无人机遥感 水稻 白叶枯病","深度学习 水稻病害 早期检测","无人机遥感水稻白叶枯病-4500","10.21203\u002Frs.3.rs-11163942\u002Fv1",{"doi":505,"openalex_id":507,"authors":508,"venue":468,"cited_by_count":150,"oa_url":466,"card":8,"direction":249,"ingested_from":169},"W7221055917",[509,512,515,517,519,522,524,526,529,531,533,535],{"name":510,"orcid":511},"Carlito B. Balingbing","https:\u002F\u002Forcid.org\u002F0000-0002-3672-9253",{"name":513,"orcid":514},"Dmytro Chebotarov","https:\u002F\u002Forcid.org\u002F0000-0003-1351-9453",{"name":516,"orcid":8},"Ian  Carlo Perilla",{"name":518,"orcid":8},"Stephen Klassen",{"name":520,"orcid":521},"Nancy Castilla","https:\u002F\u002Forcid.org\u002F0000-0002-9143-4933",{"name":523,"orcid":8},"Raphael  Gabriel Carvajal",{"name":525,"orcid":8},"Kendrix Ortega",{"name":527,"orcid":528},"Joseph M. Sandro","https:\u002F\u002Forcid.org\u002F0000-0002-8311-2299",{"name":530,"orcid":8},"Nurmi Pangesti",{"name":532,"orcid":8},"Caesar Arloo R. Centeno",{"name":534,"orcid":8},"Arnel Rala",{"name":536,"orcid":537},"Li-Yu  Daisy Liu","https:\u002F\u002Forcid.org\u002F0000-0001-6997-8101","2026-10-09T23:30:18.390549Z"]