[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-4933":3,"related-4933":61},{"id":4,"title":5,"url":6,"summary":7,"summary_zh":8,"content":8,"source_name":9,"source_url":8,"published_at":10,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":14,"score_detail":15,"sources":23,"tags":25,"primary_topic":31,"secondary_topics":37,"classification":38,"search_phrases":56,"slug":59,"view_count":21,"doi":8,"paper":8,"created_at":60},4933,"中国农科院信息所提出层次知识引导枸杞害虫识别框架","https:\u002F\u002Fnews.qq.com\u002Frain\u002Fa\u002F20261013A07CL800","中国农业科学院农业信息研究所科学数据研究室提出一种层次知识引导的枸杞害虫识别框架,有效解决真实农业场景下样本稀缺与类间混淆的双重挑战。相关研究成果发表在《人工智能的工程应用》(Engineering Applications of Artificial Intelligence)上。该研究依托长文本-图像检索模型,把专家撰写的长文本描述解析为粗、中、细三个语义层次,并与多尺度视觉特征实现精准跨模态对齐,为枸杞等特色作物的虫害智能监测与精准防控提供新的技术路径。",null,"中国农业科学院农业信息研究所 腾讯新闻 2026 年 10 月 13 日","2026-10-13T00:00:00Z","报道",10,false,78,{"impact":16,"substance":17,"depth":18,"authority":19,"freshness":20,"relevant":21,"comment":22},18,21,17,14,8,1,"国家级科研机构在农业AI领域的方法创新，跨模态对齐思路对特色作物虫害监测有推广价值，值得进入每日精选。",[24],{"name":9,"url":6},[26,27,28,29,30],"智慧农业","农业人工智能","多模态大模型","害虫识别","枸杞",{"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":13,"version":40,"method":41,"evidence":42},"classified","2026-10-22.2","rules",{"candidates":43,"agricultural_evidence":46},[44,51],{"code":32,"score":45,"supported":46,"title":47,"abstract":48,"tags":50},5,true,[],[49],"artificial intelligence",[27,28],{"code":52,"score":21,"supported":13,"title":53,"abstract":54,"tags":55},"I1",[],[],[26],[57,58],"中国农科院信息所 枸杞害虫识别","枸杞 害虫识别 视觉语言模型","中国农科院信息所枸杞害虫识别-4933","2026-10-15T00:06:01.623759Z",{"total":62,"page":21,"page_size":62,"items":63},6,[64,135,199,234,295,356],{"id":65,"title":66,"url":67,"summary":68,"summary_zh":69,"content":8,"source_name":70,"source_url":67,"published_at":71,"category":72,"cover_url":8,"hotness":12,"is_selected":13,"score":73,"score_detail":74,"sources":78,"tags":80,"primary_topic":83,"secondary_topics":84,"classification":85,"search_phrases":98,"slug":101,"view_count":102,"doi":103,"paper":104,"created_at":134},5562,"pxGPT: A schema-constrained multimodal LLM framework for plant phenotyping","https:\u002F\u002Fdoi.org\u002F10.1093\u002Fbioinformatics\u002Fbtag781","Abstract Summary Deep learning can scale image-based plant phenotyping but typically targets predefined traits, whereas manual phenotypic description is more flexible but difficult to scale. We present Phenotype eXplorer Generative Plant Traits (pxGPT), a Python framework that converts multi-view plant images into structured phenotype records. The workflow combines free-text description, dataset-specific schema generation with human review, and schema-constrained information extraction. We evaluated the extraction stage on three Brassica datasets using 78,973 annotations from ten annotators who had no access to pxGPT predictions. For 38-45 qualitative traits per dataset, pxGPT-human agreement ranged from 0.572 to 0.712, compared with 0.620 to 0.739 agreement among human annotators. All 32,736 required trait fields were present; 93 values were empty, and no qualitative descriptor violations were recorded. For selected morphological structures, pooled false-positive rates were 0.5-8.0%, while sensitivity varied across structure types in one mature-plant dataset (37.5-100%). The curated schemas were applied unchanged with two commercial models and one locally hosted open-weight model, with differences in mean accuracy of 0.03 or less. In the evaluated Brassica datasets, these results support schema-guided extraction of visible plant traits without separate model training for each trait. Human-guided schema development offers a path to extend pxGPT to other crops and imaging conditions, with accuracy requiring validation in each new setting. Availability and Implementation The software is available at https:\u002F\u002Fgithub.com\u002FYaoChengLab\u002Fpxgpt and an archival version of the source code is available on Zenodo (https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.22931745). Curated schemas used in this study, annotation guides, phenotype records and human annotations are available at https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.22907554. Supplementary information Supplementary data are available at Bioinformatics online.","摘要 深度学习能够扩展基于图像的植物表型分析，但通常针对预定义性状，而人工表型描述更为灵活却难以规模化。我们提出Phenotype eXplorer Generative Plant Traits（pxGPT），一个将多视角植物图像转换为结构化表型记录的Python框架。该工作流程结合了自由文本描述、数据集特定模式生成与人工审核，以及模式约束的信息提取。我们在三个芸薹属（Brassica）数据集上评估了提取阶段，使用了来自十名标注者的78,973条标注，这些标注者无法获取pxGPT的预测结果。对于每个数据集38-45个定性性状，pxGPT与人工标注者之间的一致性范围为0.572至0.712，而人工标注者之间的一致性为0.620至0.739。所有32,736个所需性状字段均已存在；93个值为空，未记录到定性描述符违规。对于选定的形态结构，合并假阳性率为0.5-8.0%，而在一个成熟植株数据集中，灵敏度因结构类型而异（37.5-100%）。经整理的模式在未作修改的情况下应用于两个商业模型和一个本地部署的开放权重模型，平均准确率差异不超过0.03。在所评估的芸薹属数据集中，这些结果支持在无需为每个性状单独训练模型的情况下，对可见植物性状进行模式引导的提取。人工引导的模式开发为将pxGPT扩展至其他作物和成像条件提供了途径，其准确性需在每个新设置中加以验证。可用性与实现 该软件可在https:\u002F\u002Fgithub.com\u002FYaoChengLab\u002Fpxgpt获取，源代码的存档版本可在Zenodo（https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.22931745）获取。本研究中使用的经整理模式、标注指南、表型记录和人工标注可在https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.22907554获取。补充信息 补充数据可在Bioinformatics在线获取。","Bioinformatics","2026-10-20T00:00:00Z","论文",81,{"impact":16,"substance":75,"depth":16,"authority":19,"freshness":76,"relevant":21,"comment":77},22,9,"提出基于模式约束的多模态大模型植物表型提取框架，在近8万条标注上验证，方法新颖、数据规模可观，对智慧育种与表型自动化有参考价值。",[79],{"name":70,"url":67},[26,27,81,28,82],"植物表型","芸薹属",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":13,"version":40,"method":41,"evidence":86},{"candidates":87,"agricultural_evidence":46},[88,94],{"code":32,"score":20,"supported":46,"title":89,"abstract":90,"tags":93},[],[91,92],"深度学习","deep learning",[27,28],{"code":52,"score":21,"supported":13,"title":95,"abstract":96,"tags":97},[],[],[26],[99,100],"pxGPT 植物表型","Brassica 表型 多模态大模型","pxGPT植物表型-5562",0,"10.1093\u002Fbioinformatics\u002Fbtag781",{"doi":103,"openalex_id":105,"authors":106,"venue":70,"cited_by_count":102,"oa_url":125,"card":126,"direction":132,"ingested_from":133},"W7223127960",[107,110,112,115,118,121,123],{"name":108,"orcid":109},"Po‐Xing Zheng","https:\u002F\u002Forcid.org\u002F0000-0001-5827-5865",{"name":111,"orcid":8},"Jheng-Yan Ou",{"name":113,"orcid":114},"Yu-Lin Wu","https:\u002F\u002Forcid.org\u002F0009-0003-8596-9047",{"name":116,"orcid":117},"Ruei-Shiuan Wang","https:\u002F\u002Forcid.org\u002F0009-0007-5159-1129",{"name":119,"orcid":120},"WU Ting-li","https:\u002F\u002Forcid.org\u002F0000-0001-7996-5515",{"name":122,"orcid":8},"Te-Chang Hsu",{"name":124,"orcid":8},"Yao-Cheng Lin","https:\u002F\u002Facademic.oup.com\u002Fbioinformatics\u002Fadvance-article-pdf\u002Fdoi\u002F10.1093\u002Fbioinformatics\u002Fbtag781\u002F72201337\u002Fbtag781.pdf",{"tldr":127,"method":128,"finding":129,"direction":130,"opportunity":131},"提出pxGPT框架，用多模态大模型将多视角植物图像转为结构化表型记录。","结合自由文本描述、人工审核的模式生成与模式约束信息抽取，在三个芸薹属数据集上评估","pxGPT与人工标注一致性0.572-0.712，接近人工间一致性，且无需逐性状训练模型。","农业遥感与作物表型","可探索将模式引导抽取迁移至更多作物与复杂成像条件，并研究自动化模式生成以降低人工审核成本。","智慧农业 \u002F 农业物联网","openalex","2026-10-22T11:20:22.173561Z",{"id":136,"title":137,"url":138,"summary":139,"summary_zh":140,"content":8,"source_name":141,"source_url":138,"published_at":142,"category":72,"cover_url":8,"hotness":12,"is_selected":13,"score":143,"score_detail":144,"sources":148,"tags":150,"primary_topic":153,"secondary_topics":154,"classification":155,"search_phrases":174,"slug":177,"view_count":21,"doi":178,"paper":179,"created_at":198},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",63,{"impact":145,"substance":16,"depth":146,"authority":20,"freshness":76,"relevant":21,"comment":147},12,16,"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。",[149],{"name":141,"url":138},[26,27,151,28,152],"智能体","农业决策支持",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":13,"version":40,"method":41,"evidence":156},{"candidates":157,"agricultural_evidence":46},[158,163,170],{"code":32,"score":19,"supported":46,"title":159,"abstract":161,"tags":162},[160],"decision support",[160],[27,28,151],{"code":164,"score":62,"supported":46,"title":165,"abstract":166,"tags":169},"H2",[],[167,168],"土壤","soil",[],{"code":52,"score":21,"supported":13,"title":171,"abstract":172,"tags":173},[],[],[26],[175,176],"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":178,"openalex_id":180,"authors":181,"venue":141,"cited_by_count":102,"oa_url":138,"card":192,"direction":196,"ingested_from":133},"W7222605322",[182,184,186,188,190],{"name":183,"orcid":8},"Tanmay Bhamre",{"name":185,"orcid":8},"Yash Barde",{"name":187,"orcid":8},"Devendra Yeole",{"name":189,"orcid":8},"Gajanan Chaudhary",{"name":191,"orcid":8},"Amarsinh Verpe",{"tldr":193,"method":194,"finding":195,"direction":196,"opportunity":197},"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。","2026-10-17T23:30:55.282694Z",{"id":200,"title":201,"url":202,"summary":203,"summary_zh":8,"content":204,"source_name":205,"source_url":8,"published_at":206,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":207,"score_detail":208,"sources":211,"tags":213,"primary_topic":216,"secondary_topics":217,"classification":218,"search_phrases":229,"slug":232,"view_count":45,"doi":8,"paper":8,"created_at":233},3597,"中国农科院农业信息研究所提出层次知识引导的枸杞害虫识别框架——Engineering Applications of Artificial Intelligence在线发表","https:\u002F\u002Fpaper.sciencenet.cn\u002Fhtmlnews\u002F2026\u002F9\u002F571988.shtm","中国农业科学院农业信息研究所科学数据研究室科研团队近日提出一种层次知识引导的枸杞害虫识别框架，有效解决真实农业场景下样本稀缺与类间混淆的双重挑战，相关研究成果发表在《人工智能的工程应用》（Engineering Applications of Artificial Intelligence）上。研究团队依托长文本-图像检索模型，把专家撰写的长文本描述解析为粗、中、细三个语义层次，并与多尺度视觉特征实现精准跨模态对齐。同时设计知识引导的去混淆模块，通过构建文本和视觉双模态相似度图，明显降低容易混淆类别之间的预测误差。该研究得到国家重点研发计划、中国农业科学院科技创新工程等项目支持。","[![Image 2: 科学网新闻频道](https:\u002F\u002Fpaper.sciencenet.cn\u002Fimages\u002Fnews.jpg)](https:\u002F\u002Fnews.sciencenet.cn\u002F)\n\n[生命科学](https:\u002F\u002Fwww.sciencenet.cn\u002Flife\u002F) | [医学科学](https:\u002F\u002Fwww.sciencenet.cn\u002Fmedicine\u002F) | [化学科学](https:\u002F\u002Fwww.sciencenet.cn\u002Fchemistry\u002F) | [工程材料](https:\u002F\u002Fwww.sciencenet.cn\u002Fmaterial\u002F) | [信息科学](https:\u002F\u002Fwww.sciencenet.cn\u002Finformation\u002F) | [地球科学](https:\u002F\u002Fwww.sciencenet.cn\u002Fearth\u002F) | [数理科学](https:\u002F\u002Fwww.sciencenet.cn\u002Fmathematics\u002F) | [管理综合](https:\u002F\u002Fwww.sciencenet.cn\u002Fpolicy\u002F)[站内规定](https:\u002F\u002Fblog.sciencenet.cn\u002Fblog-45-1064777.html) | [手机版](https:\u002F\u002Fwap.sciencenet.cn\u002F)\n\n[首页](https:\u002F\u002Fwww.sciencenet.cn\u002F) | [新闻](https:\u002F\u002Fnews.sciencenet.cn\u002F) | [博客](https:\u002F\u002Fblog.sciencenet.cn\u002Fblog.php) | [院士](https:\u002F\u002Fnews.sciencenet.cn\u002Fys\u002F) | [人才](https:\u002F\u002Ftalent.sciencenet.cn\u002F) | [会议](https:\u002F\u002Fmeeting.sciencenet.cn\u002F) | [基金·项目](https:\u002F\u002Ffund.sciencenet.cn\u002F) | [论文](https:\u002F\u002Fpaper.sciencenet.cn\u002F) | [绘图](https:\u002F\u002Fvisual-web.sciencenet.cn\u002F) | [视频·直播](https:\u002F\u002Fblog.sciencenet.cn\u002Fblog.php?mod=video) | [小柯机器人](https:\u002F\u002Fpaper.sciencenet.cn\u002FAInews) | [医学科普](https:\u002F\u002Fyxkp.sciencenet.cn\u002F)\n\n作者：李晨 来源：中国科学报 发布时间：2026\u002F9\u002F23 11:21:15 选择字号：小 中 大 新方法可智能识别枸杞害虫 近日，中国农业科学院农业信息研究所科学数据研究室科研团队，提出一种层次知识引导的枸杞害虫识别框架，有效解决真实农业场景下样本稀缺与类间混淆的双重挑战。相关研究成果发表在《人工智能的工程应用》（_**Engineering Applications of Artificial Intelligence**_）上。 ![Image 3](https:\u002F\u002Frmtzx.sciencenet.cn\u002Fkxwsprint\u002F6ab2a040e4b078fce44b0f9a.jpg)层次知识引导的枸杞害虫识别流程示意图。中国农科院供图 枸杞是高价值药用植物，虫害会对它的果实品质和药效造成很大影响。传统识别方法效率低且难以实现规模化应用。现有深度学习模型，面对外观高度相似的害虫种类时，常因标注样本不够，而识别效果不理想。 针对上述难题，科研团队依托长文本-图像检索模型，把专家撰写的长文本描述解析为粗、中、细三个语义层次，并与多尺度视觉特征实现精准跨模态对齐。同时，设计知识引导的去混淆模块，通过构建文本和视觉双模态相似度图，明显降低容易混淆类别之间的预测误差。该研究为枸杞等特色作物的虫害智能监测与精准防控提供新的技术路径，也为将农业专家知识融入视觉—语言模型提供参考。 该研究得到国家重点研发计划、中国农业科学院科技创新工程等项目支持。 相关论文信息：https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.engappai.2026.115342 版权声明：凡本网注明“来源：中国科学报、科学网、科学新闻杂志”的所有作品，网站转载，请在正文上方注明来源和作者，且不得对内容作实质性改动；微信公众号、头条号等新媒体平台，转载请联系授权。邮箱：shouquan@stimes.cn。\n\n打印 发E-mail给：\n\n以下评论只代表网友个人观点，不代表科学网观点。\n\n[![Image 4](https:\u002F\u002Fpaper.sciencenet.cn\u002Fimages\u002Fnewcomm.gif)](https:\u002F\u002Fpaper.sciencenet.cn\u002Fhtml\u002Fcomment.aspx?id=571988)\n\n相关新闻 相关论文\n*   1\n*   [陈文新：用后半生走遍中国“寻根”](https:\u002F\u002Fpaper.sciencenet.cn\u002Fhtmlnews\u002F2026\u002F9\u002F571931.shtm)\n\n*   2\n*   [上海市自然科学基金（启明星计划）项目申报指南发布](https:\u002F\u002Fpaper.sciencenet.cn\u002Fhtmlnews\u002F2026\u002F9\u002F571928.shtm)\n\n*   3\n*   [从实验室到生产线 高校以应用型科研对接产业升级](https:\u002F\u002Fpaper.sciencenet.cn\u002Fhtmlnews\u002F2026\u002F9\u002F571927.shtm)\n\n*   4\n*   [重点新材料研发及应用国家科技重大专项项目启动申报](https:\u002F\u002Fpaper.sciencenet.cn\u002Fhtmlnews\u002F2026\u002F9\u002F571926.shtm)\n\n*   5\n*   [一线治疗失败！晚期胃癌二线治疗有了“新标杆”](https:\u002F\u002Fpaper.sciencenet.cn\u002Fhtmlnews\u002F2026\u002F9\u002F571925.shtm)\n\n*   6\n*   [新策略助力有机室温磷光寿命大幅提升](https:\u002F\u002Fpaper.sciencenet.cn\u002Fhtmlnews\u002F2026\u002F9\u002F571909.shtm)\n\n*   7\n*   [AI绘制固态储氢材料全景导航 | 专家访谈](https:\u002F\u002Fpaper.sciencenet.cn\u002Fhtmlnews\u002F2026\u002F9\u002F571894.shtm)\n\n*   8\n*   [基金委发布1个国家自然科学基金专项项目申请指南](https:\u002F\u002Fpaper.sciencenet.cn\u002Fhtmlnews\u002F2026\u002F9\u002F571893.shtm)\n\n*   1\n*   [智能柔性传感器可原位实时监测植物体内激素](https:\u002F\u002Fpaper.sciencenet.cn\u002Fhtmlpaper\u002F2026\u002F9\u002F2026923105644661155910.shtm)\n\n*   2\n*   [浙江大学刘明团队Angew.：利用“受挫堆积”调控多孔有机笼孔道结构](https:\u002F\u002Fpaper.sciencenet.cn\u002Fhtmlpaper\u002F2026\u002F9\u002F202692310372413155909.shtm)\n\n*   3\n*   [GCB | 华南植物园揭示长期高氮沉降驱动热带森林根系分泌物加速磷活化的机制](https:\u002F\u002Fpaper.sciencenet.cn\u002Fhtmlpaper\u002F2026\u002F9\u002F2026923103553943155908.shtm)\n\n*   4\n*   [FEC | 土壤氮磷比调控高寒草地物种共存](https:\u002F\u002Fpaper.sciencenet.cn\u002Fhtmlpaper\u002F2026\u002F9\u002F2026923103451833155907.shtm)\n\n*   5\n*   [前沿聚焦：为什么烤羊肉串香？合肥工大与渤海大学联合揭示“香气密码”](https:\u002F\u002Fpaper.sciencenet.cn\u002Fhtmlpaper\u002F2026\u002F9\u002F2026923103345287155906.shtm)\n\n*   6\n*   [前沿聚焦：福州大学程树英-王伟煌联合河北大学高青：沉淀物回收策略助力 CBD 法 Sb2S3太阳电池提效](https:\u002F\u002Fpaper.sciencenet.cn\u002Fhtmlpaper\u002F2026\u002F9\u002F2026923103218818155905.shtm)\n\n*   7\n*   [Angew. Chem.：本征近红外二区荧光纳米酶用于高分辨活体成像与肿瘤催化免疫治疗](https:\u002F\u002Fpaper.sciencenet.cn\u002Fhtmlpaper\u002F2026\u002F9\u002F2026923102125755155904.shtm)\n\n*   8\n*   [2型糖尿病不同临床亚型的分子异质性被揭示](https:\u002F\u002Fpaper.sciencenet.cn\u002Fhtmlpaper\u002F2026\u002F9\u002F20269239501868155903.shtm)\n\n[图片新闻](https:\u002F\u002Fpaper.sciencenet.cn\u002Fimgnews.aspx)\n\n[![Image 5: “鹏城首发星”成功入轨](https:\u002F\u002Fnews.sciencenet.cn\u002F\u002Fupload\u002Fnews\u002Fimages\u002F2026\u002F9\u002F20269231515219900.png)](https:\u002F\u002Fnews.sciencenet.cn\u002Fhttps:\u002F\u002Fnews.sciencenet.cn\u002Fhtmlnews\u002F2026\u002F9\u002F571953.shtm)[![Image 6: 研究揭示甲烷部分氧化制合成气驱动因素](https:\u002F\u002Fnews.sciencenet.cn\u002F\u002Fupload\u002Fnews\u002Fimages\u002F2026\u002F9\u002F2026917125225680.jpg)](https:\u002F\u002Fnews.sciencenet.cn\u002F\u002Fhtmlnews\u002F2026\u002F9\u002F571610.shtm)\n[![Image 7: 科学家揭示火星大气物质运输规律](https:\u002F\u002Fnews.sciencenet.cn\u002F\u002Fupload\u002Fnews\u002Fimages\u002F2026\u002F9\u002F2026915174236470.jpg)](https:\u002F\u002Fnews.sciencenet.cn\u002F\u002Fhtmlnews\u002F2026\u002F9\u002F571503.shtm)[![Image 8: 超导双层镍氧化物薄膜的三维电子结构](https:\u002F\u002Fnews.sciencenet.cn\u002F\u002Fupload\u002Fnews\u002Fimages\u002F2026\u002F9\u002F20269151415154110.png)](https:\u002F\u002Fnews.sciencenet.cn\u002Fhttps:\u002F\u002Fnews.sciencenet.cn\u002Fhtmlnews\u002F2026\u002F9\u002F571484.shtm)\n[>>更多](https:\u002F\u002Fpaper.sciencenet.cn\u002Fimgnews.aspx)\n\n[一周新闻排行](https:\u002F\u002Fpaper.sciencenet.cn\u002Fzphnews.aspx)\n*   1\n*   [诺贝尔基金会宣布增加2026年诺奖奖金](https:\u002F\u002Fnews.sciencenet.cn\u002F\u002Fhtmlnews\u002F2026\u002F9\u002F571724.shtm)\n\n*   2\n*   [直播回放｜新型铁基催化剂，为海水制氢提速](https:\u002F\u002Fnews.sciencenet.cn\u002Fhttps:\u002F\u002Fweibo.com\u002Fl\u002Fwblive\u002Fp\u002Fshow\u002F1022:2321325343837114990618)\n\n*   3\n*   [汪品先院士：年轻人应跳出人类中心论](https:\u002F\u002Fnews.sciencenet.cn\u002F\u002Fhtmlnews\u002F2026\u002F9\u002F571776.shtm)\n\n*   4\n*   [从清华毕业做科普10年，我如何对抗AI](https:\u002F\u002Fnews.sciencenet.cn\u002F\u002Fhtmlnews\u002F2026\u002F9\u002F571671.shtm)\n\n*   5\n*   [袁慧书：脊柱“一扫多查”，让AI成为影像医生得力助手](https:\u002F\u002Fnews.sciencenet.cn\u002F\u002Fhtmlnews\u002F2026\u002F9\u002F571771.shtm)\n\n*   6\n*   [直播回放|马丁院士：共同追寻碳原子的“旅行路线”](https:\u002F\u002Fnews.sciencenet.cn\u002Fhttps:\u002F\u002Fweibo.com\u002Fl\u002Fwblive\u002Fp\u002Fshow\u002F1022:2321325344508505620608)\n\n*   7\n*   [新模型引爆AI圈：不生成文本，专为机器做快速决策](https:\u002F\u002Fnews.sciencenet.cn\u002F\u002Fhtmlnews\u002F2026\u002F9\u002F571794.shtm)\n\n*   8\n*   [AI绘制固态储氢材料全景导航 | 专家访谈](https:\u002F\u002Fnews.sciencenet.cn\u002F\u002Fhtmlnews\u002F2026\u002F9\u002F571894.shtm)\n\n*   9\n*   [AI正自主制造下一代AI？](https:\u002F\u002Fnews.sciencenet.cn\u002F\u002Fhtmlnews\u002F2026\u002F9\u002F571716.shtm)\n\n*   10\n*   [一周热闻回顾（2026年9月20日）](https:\u002F\u002Fnews.sciencenet.cn\u002F\u002Fhtmlnews\u002F2026\u002F9\u002F571820.shtm)\n\n[编辑部推荐博文](http:\u002F\u002Fblog.sciencenet.cn\u002Fblog.php?mod=recommend)\n*   ![Image 9](https:\u002F\u002Fpaper.sciencenet.cn\u002Fimages\u002Ft11.gif)\n*   [封面文章 | 效应蛋白TseMt的毒性机制及递送通路](https:\u002F\u002Fblog.sciencenet.cn\u002Fblog-45-1553810.html)\n\n*   ![Image 10](https:\u002F\u002Fpaper.sciencenet.cn\u002Fimages\u002Ft11.gif)\n*   [科学网2026年8月十佳博文榜单公布！](https:\u002F\u002Fblog.sciencenet.cn\u002Fblog-45-1551789.html)\n\n*   ![Image 11](https:\u002F\u002Fpaper.sciencenet.cn\u002Fimages\u002Ft11.gif)\n*   [我爱路亚](https:\u002F\u002Fblog.sciencenet.cn\u002Fblog-1185605-1554044.html)\n\n*   ![Image 12](https:\u002F\u002Fpaper.sciencenet.cn\u002Fimages\u002Ft11.gif)\n*   [学者的中秋家国](https:\u002F\u002Fblog.sciencenet.cn\u002Fblog-575926-1554043.html)\n\n*   ![Image 13](https:\u002F\u002Fpaper.sciencenet.cn\u002Fimages\u002Ft11.gif)\n*   [CD1 如何\"看见\"脂质](https:\u002F\u002Fblog.sciencenet.cn\u002Fblog-41174-1554027.html)\n\n*   ![Image 14](https:\u002F\u002Fpaper.sciencenet.cn\u002Fimages\u002Ft11.gif)\n*   [师傅领进门，修行在个人](https:\u002F\u002Fblog.sciencenet.cn\u002Fblog-729911-1553761.html)\n\n*   [更多>>](https:\u002F\u002Fblog.sciencenet.cn\u002Fblog.php?mod=recommend)\n\n[关于我们](http:\u002F\u002Fwww.sciencenet.cn\u002Faboutus.aspx) | [网站声明](http:\u002F\u002Fwww.sciencenet.cn\u002Fshengming.aspx) | [服务条款](http:\u002F\u002Fwww.sciencenet.cn\u002Ftiaokuang.aspx) | [联系方式](http:\u002F\u002Fwww.sciencenet.cn\u002Fcontact.aspx) | [举报](https:\u002F\u002Fwww.sciencenet.cn\u002Fjubao.aspx) | [中国科学报社](https:\u002F\u002Fstimes.sciencenet.cn\u002F)  \n[京ICP备07017567号-12](http:\u002F\u002Fwww.beian.miit.gov.cn\u002F)互联网新闻信息服务许可证10120230008 京公网安备 11010802032783  \nCopyright @ 2007-2026 中国科学报社 All Rights Reserved  \n地址：北京市海淀区中关村南一条乙三号 电话：010-62580783","科学网（中国科学报）2026-09-23","2026-09-23T00:00:00Z",77,{"impact":16,"substance":209,"depth":18,"authority":19,"freshness":20,"relevant":21,"comment":210},20,"国家级科研机构在农业AI顶刊发表的技术突破，方法新颖且直击样本稀缺与类间混淆痛点，对特色作物虫害智能监测有示范价值，值得进入每日精选。",[212],{"name":205,"url":202},[26,27,214,215,30],"病虫害识别","视觉语言模型",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":13,"version":40,"method":41,"evidence":219},{"candidates":220,"agricultural_evidence":46},[221,225],{"code":32,"score":19,"supported":46,"title":222,"abstract":223,"tags":224},[49],[49],[27,214,215],{"code":52,"score":21,"supported":13,"title":226,"abstract":227,"tags":228},[],[],[26],[230,231],"中国农科院 农业信息研究所 枸杞害虫","层次知识引导 害虫识别 视觉语言模型","中国农科院农业信息研究所枸杞害虫-3597","2026-09-27T00:05:17.078840Z",{"id":235,"title":236,"url":237,"summary":238,"summary_zh":239,"content":8,"source_name":141,"source_url":237,"published_at":240,"category":72,"cover_url":8,"hotness":12,"is_selected":13,"score":241,"score_detail":242,"sources":244,"tags":246,"primary_topic":250,"secondary_topics":256,"classification":258,"search_phrases":275,"slug":278,"view_count":102,"doi":279,"paper":280,"created_at":294},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":20,"substance":17,"depth":16,"authority":145,"freshness":12,"relevant":21,"comment":243},"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。",[245],{"name":141,"url":237},[26,27,247,248,249],"小农户","植物病害识别","可解释性",{"code":251,"slug":252,"name":253,"path":254,"parent_code":255},"G2","farm-services","社会化服务与规模经营","\u002Ftopic\u002Frural-services\u002Ffarm-services","G",[257],{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},{"status":39,"locked":13,"version":40,"method":41,"evidence":259},{"candidates":260,"agricultural_evidence":46},[261,266,271],{"code":251,"score":145,"supported":46,"title":262,"abstract":264,"tags":265},[263],"smallholder",[263],[247],{"code":32,"score":12,"supported":46,"title":267,"abstract":269,"tags":270},[268],"computer vision",[],[27,248],{"code":52,"score":21,"supported":13,"title":272,"abstract":273,"tags":274},[],[],[26],[276,277],"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":279,"openalex_id":281,"authors":282,"venue":141,"cited_by_count":102,"oa_url":237,"card":289,"direction":196,"ingested_from":133},"W7223137323",[283,285,287],{"name":284,"orcid":8},"Akshay Ramgoolam",{"name":286,"orcid":8},"Priya Bundhoo Appadoo",{"name":288,"orcid":8},"Jean-Noël LEspérance",{"tldr":290,"method":291,"finding":292,"direction":196,"opportunity":293},"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":296,"title":297,"url":298,"summary":299,"summary_zh":300,"content":8,"source_name":141,"source_url":298,"published_at":240,"category":72,"cover_url":8,"hotness":12,"is_selected":13,"score":241,"score_detail":301,"sources":303,"tags":305,"primary_topic":309,"secondary_topics":315,"classification":317,"search_phrases":338,"slug":341,"view_count":102,"doi":342,"paper":343,"created_at":355},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":20,"substance":75,"depth":16,"authority":145,"freshness":76,"relevant":21,"comment":302},"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。",[304],{"name":141,"url":298},[26,27,306,307,308],"灌溉决策","安全保证","试验设计",{"code":310,"slug":311,"name":312,"path":313,"parent_code":314},"H3","irrigation","水资源与灌溉","\u002Ftopic\u002Fenvironment\u002Firrigation","H",[316],{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},{"status":39,"locked":13,"version":40,"method":41,"evidence":318},{"candidates":319,"agricultural_evidence":46},[320,326,330,334],{"code":310,"score":321,"supported":46,"title":322,"abstract":323,"tags":325},15,[311],[324,311],"灌溉",[306],{"code":32,"score":145,"supported":46,"title":327,"abstract":328,"tags":329},[49],[160],[27],{"code":164,"score":62,"supported":46,"title":331,"abstract":332,"tags":333},[],[167,168],[],{"code":52,"score":21,"supported":13,"title":335,"abstract":336,"tags":337},[],[],[26],[339,340],"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":342,"openalex_id":344,"authors":345,"venue":141,"cited_by_count":102,"oa_url":298,"card":350,"direction":196,"ingested_from":133},"W7223131205",[346,348],{"name":347,"orcid":8},"Nassima Benkhelifa",{"name":349,"orcid":8},"Yacine Aït Ahmed",{"tldr":351,"method":352,"finding":353,"direction":196,"opportunity":354},"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",{"id":357,"title":358,"url":359,"summary":360,"summary_zh":361,"content":8,"source_name":141,"source_url":359,"published_at":240,"category":72,"cover_url":8,"hotness":12,"is_selected":13,"score":362,"score_detail":363,"sources":365,"tags":367,"primary_topic":370,"secondary_topics":371,"classification":373,"search_phrases":388,"slug":391,"view_count":102,"doi":392,"paper":393,"created_at":407},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":20,"substance":75,"depth":16,"authority":145,"freshness":12,"relevant":21,"comment":364},"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。",[366],{"name":141,"url":359},[26,27,247,368,369],"作物病害识别","模型审计",{"code":251,"slug":252,"name":253,"path":254,"parent_code":255},[372],{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},{"status":39,"locked":13,"version":40,"method":41,"evidence":374},{"candidates":375,"agricultural_evidence":46},[376,380,384],{"code":251,"score":321,"supported":46,"title":377,"abstract":378,"tags":379},[263],[247,263],[247],{"code":32,"score":12,"supported":46,"title":381,"abstract":382,"tags":383},[268],[],[368,27],{"code":52,"score":21,"supported":13,"title":385,"abstract":386,"tags":387},[],[],[26],[389,390],"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":392,"openalex_id":394,"authors":395,"venue":141,"cited_by_count":102,"oa_url":359,"card":402,"direction":196,"ingested_from":133},"W7223088030",[396,398,400],{"name":397,"orcid":8},"Abdi-Basid Idriss Farah",{"name":399,"orcid":8},"Amina Abdillahi Guedi",{"name":401,"orcid":8},"Moussa Omar Youssouf",{"tldr":403,"method":404,"finding":405,"direction":196,"opportunity":406},"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"]