发表机构
University of Nebraska-Lincoln; Panhandle Research and Extension Center（内布拉斯加大学林肯分校; 狭长地带研究与推广中心）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究针对精准杂草控制问题，提出通过可验证奖励学习视觉基础植物推理的WeedExpert-R1模型，利用特定合成管道生成数据微调，经群体相对策略优化训练。该模型在多种杂草测试中表现优异，优于同类模型，展现开放词汇能力和跨区域部署潜力。
AI 中文摘要
精准杂草控制需要物种级识别和实例级定位。传统目标检测器存在局限性，多模态大语言模型虽有视觉基础和推理能力，但植物知识不足。本研究引入WeedExpert-R1，通过可验证奖励学习视觉基础植物推理。利用特定领域思维链合成管道生成推理数据进行监督微调，应用群体相对策略优化。在六个数据集的37种杂草上测试，WeedExpert-R1-4B在IoU阈值为0.5时精确集精度达75.82%，精度89.30%，召回率87.81%，优于专有和开源模型，对未见物种的结果展示其开放词汇能力和跨区域部署潜力。
英文摘要
Precision weed control requires species-level identification and instance-level localization. However, conventional object detectors use a closed vocabulary, limiting their deployment across regions, and cannot explain their predictions in complex agricultural scenes. Multimodal large language models (MLLMs) offer visual grounding and reasoning capabilities, but insufficient botanical knowledge can cause hallucinations in fine-grained weed identification. This study introduces WeedExpert-R1, a multimodal model that learns visually grounded botanical reasoning through verifiable rewards. A domain-specific Chain-of-Thought synthesis pipeline combines a human-curated botanical trait dictionary with an Auditor-Synthesizer LLM workflow to generate reasoning data for supervised fine-tuning. Group Relative Policy Optimization is then applied with rewards for format, accuracy, instance count, and response length. Across 37 weed species from six datasets, WeedExpert-R1-4B achieved 75.82 percent exact-set precision at an IoU threshold of 0.5, 89.30 percent precision, and 87.81 percent recall. It outperformed proprietary models, including GPT-5.4 and Gemini-3.1-Pro, and larger open-source models, including Qwen3-VL-30B-Instruct and Gemma-4-31B-it. Results on unseen species further demonstrate its open-vocabulary capability and potential for deployment across diverse regions and crops without retraining.
