{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,21]],"date-time":"2026-09-21T06:02:14Z","timestamp":1789970534839,"version":"4.0.1"},"reference-count":45,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,21]],"date-time":"2026-09-21T00:00:00Z","timestamp":1789948800000},"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. Plant Sci."],"abstract":"<jats:p>In plant disease classification tasks, existing task specific models for plant disease classification typically achieve high accuracy by employing tailored network architectures. However, these models are unable to provide reasoning chains, which reduces interpretability and undermines trustworthiness in practical applications. In addition, although multimodal large language models (MLLMs) have made significant progress in smart agriculture, their disease classification accuracy lags behind that of task-specific models due to insufficient domain knowledge. While fine-tuning can enhance domain-specific knowledge, relying solely on unimodal data or simple multimodal information remains insufficient for adequately improving classification accuracy. Specifically, the simple multimodal information in this context refers to a single global description of the entire image produced by an MLLM. To address these issues, this paper proposes a multi-agent collaborative classification framework that integrates Monte Carlo Tree Search (MCTS) with multimodal enhancement. With an MLLM as the core, the framework first introduces an MCTSguided multi-agent collaboration mechanism that directs the agents to progressively generate region-specific descriptive textual information, which together with the image forms complex multimodal information and thereby boosts classification performance. Meanwhile, the search trajectories generated by MCTS are leveraged to construct reasoning processes for enhanced interpretability. Finally, a self-training mechanism is adopted to iteratively optimize the model, further improving overall performance. Experimental results show that the framework improves the baseline MLLM\u2019s accuracy while providing interpretable reasoning chains, achieving gains from 0.6701 to 0.9965 on 3-class potato, 0.5340 to 0.9651 on 5-class potato, 0.5387 to 0.9467 on tomato, and 0.4705 to 0.9948 on grape.<\/jats:p>","DOI":"10.3389\/fpls.2026.1937679","type":"journal-article","created":{"date-parts":[[2026,9,21]],"date-time":"2026-09-21T05:36:42Z","timestamp":1789969002000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["A multi-agent multi-modal LLM with Monte Carlo tree search for interpretable plant disease classification"],"prefix":"10.3389","volume":"17","author":[{"given":"Ridong","family":"Wu","sequence":"first","affiliation":[{"name":"School of Computer and Information Engineering, Xi\u2019an University of Architecture and Technology","place":["Xi\u2019an, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hua","family":"Zou","sequence":"additional","affiliation":[{"name":"School of Computer Science, Wuhan University","place":["Wuhan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhijie","family":"Li","sequence":"additional","affiliation":[{"name":"School of Computer and Information Engineering, Xi\u2019an University of Architecture and Technology","place":["Xi\u2019an, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Qian","family":"Zhou","sequence":"additional","affiliation":[{"name":"School of Computer Science, Wuhan University","place":["Wuhan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zijiang","family":"Yi","sequence":"additional","affiliation":[{"name":"School of Computer and Information Engineering, Xi\u2019an University of Architecture and Technology","place":["Xi\u2019an, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,21]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"1516","DOI":"10.3390\/math11061516","article-title":"Potato blight detection using fine-tuned cnn architecture","volume":"11","author":"Al-Adhaileh","year":"2023","journal-title":"Mathematics"},{"key":"B2","doi-asserted-by":"publisher","first-page":"406","DOI":"10.1016\/j.aej.2023.07.076","article-title":"Pldpnet: End-to-end hybrid deep learning framework for potato leaf disease prediction","volume":"78","author":"Arshad","year":"2023","journal-title":"Alexandria Eng. J."},{"key":"B3","doi-asserted-by":"publisher","DOI":"10.3389\/fpls.2024.1505857","article-title":"Enhancing plant disease detection through deep learning: a depthwise cnn with squeeze and excitation integration and residual skip connections","volume":"15","author":"Ashurov","year":"2025","journal-title":"Front. Plant Sci."},{"key":"B4","article-title":"Qwen3-vl technical report","volume-title":"Arxiv Preprint Arxiv:2511.21631","author":"Bai","year":"2025"},{"key":"B5","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TCIAIG.2012.2186810","article-title":"A survey of monte carlo tree search methods","volume":"4","author":"Browne","year":"2012","journal-title":"IEEE Trans. Comput. Intell. AI Games"},{"key":"B6","doi-asserted-by":"publisher","DOI":"10.1016\/j.compag.2023.107993","article-title":"Cucumber disease recognition with small samples using image-text-label-based multi-modal language model","volume":"211","author":"Cao","year":"2023","journal-title":"Comput. Electron. Agric."},{"key":"B7","doi-asserted-by":"publisher","first-page":"3227","DOI":"10.1007\/s11540-025-09877-7","article-title":"Mish-cnn: An enhanced deep learning model with mish activation function for potato leaf disease detection","volume":"68","author":"Chaibi","year":"2025","journal-title":"Potato Res."},{"key":"B8","doi-asserted-by":"publisher","DOI":"10.1016\/j.pmpp.2021.101781","article-title":"Automated recognition of optical image based potato leaf blight diseases using deep learning","volume":"117","author":"Chakraborty","year":"2022","journal-title":"Physiol. Mol. Plant Pathol."},{"key":"B9","first-page":"2607","article-title":"Main-rag: Multi-agent filtering retrieval-augmented generation","author":"Chang","year":"2025"},{"key":"B10","doi-asserted-by":"publisher","first-page":"1031","DOI":"10.3390\/agriculture16101031","article-title":"Hdr-yolo: A lightweight tomato leaf disease detection model based on high-level semantic guidance and dynamic upsampling","volume":"16","author":"Chen","year":"2026","journal-title":"Agriculture"},{"key":"B11","doi-asserted-by":"publisher","DOI":"10.1016\/j.compag.2023.108481","article-title":"Dfn-psan: Multi-level deep information feature fusion extraction network for interpretable plant disease classification","volume":"216","author":"Dai","year":"2024","journal-title":"Comput. Electron. Agric."},{"key":"B12","doi-asserted-by":"publisher","DOI":"10.1016\/j.engappai.2025.110743","article-title":"Semi-supervised deep learning framework based on modified pyramid scene parsing network for multi-label fine-grained classification and diagnosis of apple leaf diseases","volume":"151","author":"Fan","year":"2025","journal-title":"Eng. Appl. Artif. Intell."},{"key":"B13","first-page":"6577","article-title":"A survey of confidence estimation and calibration in large language models","author":"Geng","year":"2024"},{"key":"B14","doi-asserted-by":"publisher","DOI":"10.3390\/electronics13030494","article-title":"Multi-agent vision system for supporting autonomous orchard spraying","volume":"13","author":"G\u00f3ral","year":"2024","journal-title":"Electronics"},{"key":"B15","doi-asserted-by":"publisher","DOI":"10.3390\/agriengineering7110394","article-title":"Ai-driven cooperative control for autonomous tractors and implements: A comprehensive review","volume":"7","author":"Jia","year":"2025","journal-title":"AgriEngineering"},{"key":"B16","first-page":"14674","article-title":"WISE: Weak-supervisionguided step-by-step explanations for multimodal LLMs in image classification","author":"Jiang","year":"2025"},{"key":"B17","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s42269-021-00627-6","article-title":"Identification of plant diseases and distinct approaches for their management","volume":"45","author":"Kaur","year":"2021","journal-title":"Bull. Natl. Res. Cent."},{"key":"B18","doi-asserted-by":"publisher","DOI":"10.36227\/techrxiv.175979217.76947425\/v1","article-title":"The describe-then-generate bottleneck: How vlm descriptions alter image generation outcomes","author":"Kodathala","year":"2025"},{"key":"B19","first-page":"801","article-title":"Ai powered smart recognition of plant diseases with vernacular language support","author":"Kumar","year":"2025"},{"key":"B20","doi-asserted-by":"publisher","first-page":"303","DOI":"10.1080\/10807039.2022.2064814","article-title":"A novel framework for potato leaf disease detection using an efficient deep learning model","volume":"29","author":"Mahum","year":"2023","journal-title":"Hum. Ecol. Risk Assess."},{"key":"B21","doi-asserted-by":"publisher","DOI":"10.3389\/fpls.2016.01419","article-title":"Using deep learning for image-based plant disease detection","volume":"7","author":"Mohanty","year":"2016","journal-title":"Front. Plant Sci."},{"key":"B22","first-page":"4384","article-title":"Building multi-agent copilot towards autonomous agricultural data management and analysis","author":"Pan","year":"2024"},{"key":"B23","first-page":"1","article-title":"Odnet: Optimized deep convolutional neural network for classification of solanum tuberosum leaves diseases","author":"Pandiri","year":"2022"},{"key":"B24","first-page":"1271","article-title":"Pdd-agent: Multimodal large language model-driven ai agent for enhanced plant disease diagnosis","author":"Qin","year":"2025"},{"key":"B25","doi-asserted-by":"publisher","DOI":"10.3390\/electronics10172064","article-title":"Multi-level deep learning model for potato leaf disease recognition","volume":"10","author":"Rashid","year":"2021","journal-title":"Electronics"},{"key":"B26","doi-asserted-by":"publisher","DOI":"10.1016\/j.engappai.2024.108307","article-title":"Potato leaf disease detection with a novel deep learning model based on depthwise separable convolution and transformer networks","volume":"133","author":"Reis","year":"2024","journal-title":"Eng. Appl. Artif. Intel."},{"key":"B27","doi-asserted-by":"publisher","DOI":"10.1016\/j.asoc.2023.110534","article-title":"Leaf disease detection using machine learning and deep learning: review and challenges","volume":"145","author":"Sarkar","year":"2023","journal-title":"Appl. Soft Comput."},{"key":"B28","doi-asserted-by":"publisher","DOI":"10.3390\/s23239516","article-title":"Efficientrmt-net\u2014an efficient resnet-50 and vision transformers approach for classifying potato plant leaf diseases","volume":"23","author":"Shaheed","year":"2023","journal-title":"Sensors"},{"key":"B29","first-page":"392","article-title":"Potato leaf disease classification using deep learning approach","author":"Sholihati","year":"2020"},{"key":"B30","doi-asserted-by":"publisher","first-page":"354","DOI":"10.1038\/nature24270","article-title":"Mastering the game of go without human knowledge","volume":"550","author":"Silver","year":"2017","journal-title":"Nature"},{"key":"B31","first-page":"9288","article-title":"Rekg-mcts: Reinforcing llm reasoning on knowledge graphs via training-free monte carlo tree search","author":"Song","year":"2025"},{"key":"B32","article-title":"Litecoop: Lightweight multi-llm shared-tree reasoning for model-serving compiler optimizations","volume-title":"Arxiv:2602.01935 [Cs.LG]","author":"Tang","year":"2024"},{"key":"B33","doi-asserted-by":"publisher","DOI":"10.1016\/j.engappai.2025.112661","article-title":"Optimizing agricultural order fulfillment systems: A hybrid tree search approach","volume":"162","author":"Thangeda","year":"2025","journal-title":"Eng. Appl. Artif. Intell."},{"key":"B34","first-page":"18305","article-title":"Rare: Retrieval-augmented reasoning enhancement for large language models","author":"Tran","year":"2025"},{"key":"B35","doi-asserted-by":"publisher","first-page":"21959","DOI":"10.1038\/s41598-025-01908-0","article-title":"A large language model for multimodal identification of crop diseases and pests","volume":"15","author":"Wang","year":"2025","journal-title":"Sci. Rep."},{"key":"B36","doi-asserted-by":"crossref","DOI":"10.52202\/085713-2193","article-title":"Concept-guided interpretability via neural chunking","volume-title":"Arxiv Preprint Arxiv:2505.11576","author":"Wu","year":"2025"},{"key":"B37","doi-asserted-by":"publisher","DOI":"10.1016\/j.plaphe.2025.100132","article-title":"Plantif: Multimodal semantic interactive fusion via graph learning for plant disease diagnosis","volume":"8","author":"Wu","year":"2026","journal-title":"Plant Phenomics"},{"key":"B38","first-page":"8678","article-title":"Sra-mcts: Self-driven reasoning augmentation with monte carlo tree search for code generation","author":"Xu","year":"2025"},{"key":"B39","first-page":"15474","article-title":"Pride and prejudice: LLM amplifies selfbias in self-refinement","author":"Xu","year":"2024"},{"key":"B40","doi-asserted-by":"publisher","DOI":"10.1016\/j.engappai.2025.111665","article-title":"A multi-agent-based cooperative optimization control method for the motor energy consumption and tracking accuracy of an unmanned electric tractor","volume":"159","author":"Xu","year":"2025","journal-title":"Eng. Appl. Artif. Intell."},{"key":"B41","first-page":"14358","article-title":"Rod-mllm: Towards more reliable object detection in multimodal large language models","author":"Yin","year":"2025"},{"key":"B42","first-page":"1444","article-title":"Empowering large language model agent through step-level self-critique and self-training","author":"Zhai","year":"2025"},{"key":"B43","article-title":"Next-chat: An lmm for chat, detection and segmentation","volume-title":"Arxiv Preprint Arxiv:2311.04498","author":"Zhang","year":"2023"},{"key":"B44","doi-asserted-by":"publisher","DOI":"10.1016\/j.compag.2021.106408","article-title":"Crop disease identification and interpretation method based on multimodal deep learning","volume":"189","author":"Zhou","year":"2021","journal-title":"Comput. Electron. Agric."},{"key":"B45","doi-asserted-by":"publisher","DOI":"10.1016\/j.compag.2024.109824","article-title":"Potato disease detection and prevention using multimodal ai and large language model","volume":"229","author":"Zhu","year":"2025","journal-title":"Comput. Electron. Agric."}],"container-title":["Frontiers in Plant Science"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpls.2026.1937679\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,21]],"date-time":"2026-09-21T05:36:43Z","timestamp":1789969003000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpls.2026.1937679\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,21]]},"references-count":45,"alternative-id":["10.3389\/fpls.2026.1937679"],"URL":"https:\/\/doi.org\/10.3389\/fpls.2026.1937679","relation":{},"ISSN":["1664-462X"],"issn-type":[{"value":"1664-462X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,21]]},"article-number":"1937679"}}