{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,13]],"date-time":"2026-10-13T17:12:48Z","timestamp":1791911568468,"version":"4.3.4"},"reference-count":31,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,17]],"date-time":"2026-09-17T00:00:00Z","timestamp":1789603200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Artif. Intell."],"abstract":"<jats:p>\n                    Accurate crop yield forecasting in commercial soft fruit production is constrained by the data available in typical commercial farm records, which lack the sensor networks, satellite imagery, and high-resolution meteorological inputs that most state-of-the-art approaches assume. Even when provided with recent historical observations, existing forecasting models exhibit systematic, season-specific errors that growers can readily identify but that the models have no mechanism to self-correct. This work proposes a structured LLM agent framework that performs\n                    <jats:italic>post-hoc<\/jats:italic>\n                    correction of existing model predictions without requiring additional data or retraining, encoding agricultural domain knowledge across a five-tool pipeline covering seasonal phase detection, bias learning, range validation, cross-plot similarity search, and correction synthesis. The framework operates as a model-agnostic layer on top of any base forecaster, with each correction traceable to an explicit reasoning step. Evaluated on a proprietary strawberry yield dataset and a public USDA corn harvest dataset, agent refinement of XGBoost reduced MAE by 22% and MASE by 54% on strawberry, with consistent improvements across Moirai2 (MAE \u221225%, MASE \u221222%) and Random Forest (MAE \u221231%, MASE \u221261%) baselines. Using Llama 3.1 8B as the agent produced the strongest LLM-based corrections across all configurations, the Deterministic baseline was competitive with or outperformed Llama 3.1 8B on XGBoost and Random Forest, suggesting that domain-tool correction alone recovers much of the available gain. LLaVA 13B showed inconsistent gains, highlighting sensitivity to the choice of refinement model.\n                  <\/jats:p>","DOI":"10.3389\/frai.2026.1911884","type":"journal-article","created":{"date-parts":[[2026,9,17]],"date-time":"2026-09-17T05:31:30Z","timestamp":1789623090000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["Agent-based post-hoc correction of agricultural yield forecasts"],"prefix":"10.3389","volume":"9","author":[{"given":"Matthew","family":"Beddows","sequence":"first","affiliation":[{"name":"School of Natural and Computing Sciences, University of Aberdeen","place":["Aberdeen, United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Aiden","family":"Durrant","sequence":"additional","affiliation":[{"name":"School of Computing Sciences, University of East Anglia","place":["Norwich, United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Georgios","family":"Leontidis","sequence":"additional","affiliation":[{"name":"Department of Physics and Technology, UiT The Arctic University of Norway","place":["Troms\u00f8, Norway"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,17]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"425","DOI":"10.17660\/ActaHortic.2020.1296.55","article-title":"\u201cUsing deep learning to predict plant growth and yield in greenhouse environments,\u201d","author":"Alhnaity","year":"2019","journal-title":"International symposium on advanced technologies and management for innovative greenhouses: GreenSys2019, Vol. 1296"},{"key":"B2","doi-asserted-by":"publisher","first-page":"e039676","DOI":"10.1136\/bmjopen-2020-039676","article-title":"Comparison of ARIMA model and XGBoost model for prediction of human brucellosis in mainland China: a time-series study","volume":"10","author":"Alim","year":"2020","journal-title":"BMJ Open"},{"key":"B3","doi-asserted-by":"publisher","first-page":"2096","DOI":"10.3390\/app15042096","article-title":"AI algorithms in the agrifood industry: application potential in the Spanish agrifood context","volume":"15","author":"Ar\u00e9valo-Royo","year":"2025","journal-title":"Appl. Sci"},{"key":"B4","doi-asserted-by":"publisher","first-page":"626","DOI":"10.1002\/ppp3.70110","article-title":"Protected and productive: how greenhouses should deliver UK food security","volume":"8","author":"Batke","year":"2025","journal-title":"Plants People Planet"},{"key":"B5","doi-asserted-by":"publisher","first-page":"128","DOI":"10.2139\/ssrn.5202498","article-title":"VisionTreeS: a hybrid tree-based visual masked autoencoder approach for strawberry yield forecasting from low-resolution data","volume":"4","author":"Beddows","year":"2025","journal-title":"IEEE Trans. AgriFood Electron"},{"key":"B6","doi-asserted-by":"publisher","first-page":"883","DOI":"10.3390\/agriculture14060883","article-title":"A multi-farm global-to-local expert-informed machine learning system for strawberry yield forecasting","volume":"14","author":"Beddows","year":"2024","journal-title":"Agriculture"},{"key":"B7","doi-asserted-by":"publisher","first-page":"139","DOI":"10.1007\/978-3-031-63793-3_8","article-title":"\u201cA review of yield forecasting techniques and their impact on sustainable agriculture,\u201d","author":"Celis","year":"2024","journal-title":"Transformation towards Circular Food Systems"},{"key":"B8","article-title":"Deep learning approaches for forecasting strawberry yields and prices using satellite images and station-based soil parameters","author":"Chaudhary","year":"2021","journal-title":"arXiv"},{"key":"B9","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3698589","article-title":"Empowering agrifood system with artificial intelligence: a survey of the progress, challenges and opportunities","volume":"57","author":"Chen","year":"2024","journal-title":"ACM Comput. Surv"},{"key":"B10","doi-asserted-by":"publisher","DOI":"10.2139\/ssrn.4464002","article-title":"Chatgpt informed graph neural network for stock movement prediction","author":"Chen","year":"2023","journal-title":"arXiv"},{"key":"B11","doi-asserted-by":"publisher","first-page":"1852","DOI":"10.1108\/BFJ-06-2022-0490","article-title":"Technological innovation in agri-food supply chains","volume":"126","author":"Cricelli","year":"2024","journal-title":"Br. Food J"},{"key":"B12","doi-asserted-by":"publisher","first-page":"106648","DOI":"10.1016\/j.compag.2021.106648","article-title":"The role of cross-silo federated learning in facilitating data sharing in the agri-food sector","volume":"193","author":"Durrant","year":"2022","journal-title":"Comput. Electron. Agric"},{"key":"B13","article-title":"\u201cEmpowering time series analysis with large language models: a survey,\u201d","volume-title":"33rd international joint conference on artificial intelligence (IJCAI 2024)","author":"Jin","year":"2024"},{"key":"B14","article-title":"Time-LLM: time series forecasting by reprogramming large language models","author":"Jin","year":"2023","journal-title":"arXiv"},{"key":"B15","doi-asserted-by":"publisher","first-page":"105103","DOI":"10.1016\/j.compag.2019.105103","article-title":"A framework for predicting soft-fruit yields and phenology using embedded, networked microsensors, coupled weather models and machine-learning techniques","volume":"168","author":"Lee","year":"2020","journal-title":"Comput. Electron. Agric"},{"key":"B16","doi-asserted-by":"publisher","first-page":"119","DOI":"10.1145\/3643795.3648387","article-title":"\u201cLLMS for relational reasoning: how far are we?\u201d","author":"Li","year":"2024","journal-title":"Proceedings of the 1st international workshop on large language models for code"},{"key":"B17","doi-asserted-by":"publisher","first-page":"5774","DOI":"10.1109\/ICCV51070.2023.00531","article-title":"\u201cMMST-ViT: climate change-aware crop yield prediction via multi-modal spatial-temporal vision transformer,\u201d","author":"Lin","year":"2023","journal-title":"Proceedings of the IEEE\/CVF international conference on computer vision"},{"key":"B18","article-title":"Moirai 2.0: when less is more for time series forecasting","author":"Liu","year":"2025","journal-title":"arXiv"},{"key":"B19","doi-asserted-by":"publisher","first-page":"110212","DOI":"10.1016\/j.compag.2025.110212","article-title":"AI-driven time series analysis for predicting strawberry weekly yields integrating fruit monitoring and weather data for optimized harvest planning","volume":"233","author":"Liu","year":"2025","journal-title":"Comput. Electron. Agric"},{"key":"B20","doi-asserted-by":"publisher","first-page":"111564","DOI":"10.1016\/j.compag.2026.111564","article-title":"NutriCHAT: a reasoning-driven large language model agent with expert-designed tools for knowledge-grounded poultry nutrition assistance","volume":"245","author":"Mandiga","year":"2026","journal-title":"Comput. Electron. Agric"},{"key":"B21","doi-asserted-by":"publisher","first-page":"7405","DOI":"10.3390\/app13137405","article-title":"Artificial intelligence in agriculture: benefits, challenges, and trends","volume":"13","author":"Oliveira","year":"2023","journal-title":"Appl. Sci"},{"key":"B22","doi-asserted-by":"publisher","first-page":"107784","DOI":"10.1016\/j.compag.2023.107784","article-title":"Premonition net, a multi-timeline transformer network architecture towards strawberry tabletop yield forecasting","volume":"208","author":"Onoufriou","year":"2023","journal-title":"Comput. Electron. Agric"},{"key":"B23","doi-asserted-by":"publisher","first-page":"195","DOI":"10.1007\/s41324-020-00346-6","article-title":"Crop yield prediction in cotton for regional level using random forest approach","volume":"29","author":"Prasad","year":"2021","journal-title":"Spat. Inf. Res"},{"key":"B24","doi-asserted-by":"publisher","first-page":"570","DOI":"10.1007\/s11119-023-10083-z","article-title":"Forecasting carrot yield with optimal timing of sentinel 2 image acquisition","volume":"25","author":"Suarez","year":"2024","journal-title":"Precis. Agric"},{"key":"B25","doi-asserted-by":"publisher","first-page":"7105","DOI":"10.1038\/s41598-025-91450-w","article-title":"Precision biochar yield forecasting employing random forest and XGBoost with Taylor diagram visualization","volume":"15","author":"Uppalapati","year":"2025","journal-title":"Sci. Rep"},{"key":"B26","article-title":"\u201cUnified training of universal time series forecasting transformers,\u201d","volume-title":"Proceedings of the 41st international conference on machine learning","author":"Woo","year":"2024"},{"key":"B27","doi-asserted-by":"publisher","first-page":"338","DOI":"10.1109\/MGRS.2025.3571906","article-title":"Progress and perspectives of crop yield forecasting with remote sensing: a review","volume":"13","author":"Xiao","year":"2025","journal-title":"IEEE Geosci. Remote Sens. Mag"},{"key":"B28","article-title":"\u201cReAct: synergizing reasoning and acting in language models,\u201d","volume-title":"The eleventh international conference on learning representations","author":"Yao","year":"2022"},{"key":"B29","doi-asserted-by":"publisher","DOI":"10.1109\/BigData66926.2025.11402044","article-title":"Empowering time series forecasting with LLM-agents","author":"Yeh","year":"2025","journal-title":"arXiv"},{"key":"B30","doi-asserted-by":"publisher","first-page":"739","DOI":"10.18653\/v1\/2023.emnlp-industry.69","article-title":"\u201cHarnessing LLMS for temporal data-a study on explainable financial time series forecasting,\u201d","author":"Yu","year":"2023","journal-title":"Proceedings of the 2023 conference on empirical methods in natural language processing: industry track"},{"key":"B31","doi-asserted-by":"publisher","first-page":"4511","DOI":"10.3390\/rs14184511","article-title":"Prediction of strawberry dry biomass from UAV multispectral imagery using multiple machine learning methods","volume":"14","author":"Zheng","year":"2022","journal-title":"Remote Sens"}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1911884\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,17]],"date-time":"2026-09-17T05:31:32Z","timestamp":1789623092000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1911884\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,17]]},"references-count":31,"alternative-id":["10.3389\/frai.2026.1911884"],"URL":"https:\/\/doi.org\/10.3389\/frai.2026.1911884","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,17]]},"article-number":"1911884"}}