arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2510.04002cs.CL
收藏

AgriGPT-VL:农业视觉-语言理解套件

AgriGPT-VL: Agricultural Vision-Language Understanding Suite

Zhejiang University(浙江大学)

Bo Yang, Yunkui Chen, Lanfei Feng, Yu Zhang, Xiao Xu, Jianyu Zhang, Nueraili Aierken, Runhe Huang, Hongjian Lin, Yibin Ying, Shijian Li

AI总结:

AgriGPT-VL通过构建农业专用的视觉-语言模型和评估套件,提升了农业领域的多模态理解和推理能力。

摘要:

尽管多模态大语言模型取得了快速进展,但农业应用仍受到领域定制模型稀缺、精心编纂的视觉-语言语料库匮乏以及严格评估的限制。为解决这些挑战,我们提出了AgriGPT-VL套件,一个统一的多模态框架用于农业。我们的贡献有三方面。首先,我们介绍了Agri-3M-VL,据我们所知,这是最大的农业视觉-语言语料库,由可扩展的多代理数据生成器编纂;它包含100万对图像-描述词对,200万对图像基础VQA对,5万例专家级VQA实例,以及1.5万例GRPO强化学习样本。其次,我们开发了AgriGPT-VL,一个专门用于农业的视觉-语言模型,通过文本基础的渐进课程、多模态浅层/深层对齐以及GRPO优化进行训练。该方法在实现强大的多模态推理的同时,仍保留了纯文本能力。第三,我们建立了AgriBench-VL-4K,一个紧凑且具有挑战性的评估套件,包含开放性和图像基础的问题,并配以多指标评估和LLM-as-a-judge框架。实验显示,AgriGPT-VL在AgriBench-VL-4K上优于领先的通用VLMs,实现了更高的LLM-as-a-judge评估中的 pairwise win rates。同时,它在文本-only的AgriBench-13K上仍具有竞争力,没有明显降级语言能力。消融研究进一步证实了我们对齐和GRPO优化阶段的一致收益。我们将开源所有资源以支持可重复的研究和在低资源农业环境中的部署。

英文摘要:

Despite rapid advances in multimodal large language models, agricultural applications remain constrained by the scarcity of domain-tailored models, curated vision-language corpora, and rigorous evaluation. To address these challenges, we present the AgriGPT-VL Suite, a unified multimodal framework for agriculture. Our contributions are threefold. First, we introduce Agri-3M-VL, the largest vision-language corpus for agriculture to our knowledge, curated by a scalable multi-agent data generator; it comprises 1M image-caption pairs, 2M image-grounded VQA pairs, 50K expert-level VQA instances, and 15K GRPO reinforcement learning samples. Second, we develop AgriGPT-VL, an agriculture-specialized vision-language model trained via a progressive curriculum of textual grounding, multimodal shallow/deep alignment, and GRPO refinement. This method achieves strong multimodal reasoning while preserving text-only capability. Third, we establish AgriBench-VL-4K, a compact yet challenging evaluation suite with open-ended and image-grounded questions, paired with multi-metric evaluation and an LLM-as-a-judge framework. Experiments show that AgriGPT-VL outperforms leading general-purpose VLMs on AgriBench-VL-4K, achieving higher pairwise win rates in the LLM-as-a-judge evaluation. Meanwhile, it remains competitive on the text-only AgriBench-13K with no noticeable degradation of language ability. Ablation studies further confirm consistent gains from our alignment and GRPO refinement stages. We will open source all of the resources to support reproducible research and deployment in low-resource agricultural settings.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑