Skip to content

Dataset · risashinoda

Agricultural multi-modal model benchmark

risashinoda/AgroBench

Image-text dataset from ICCV '25 for evaluating vision-language models in agriculture.

What's inside

The dataset contains between 1,000 and 10,000 image-text pairs collected by research authors for ICCV 2025. It is designed to test multi-modal models on agricultural tasks.

What you can do with it

  • Evaluating vision-language model (VLM) accuracy in agro-industrial tasks
  • Testing plant disease recognition and crop analysis photo systems
  • Validating multi-modal assistants for farmers and agronomists
Size
1K–10K rows
Downloads this month
4.8K
Likes
27
Access
on request on Hugging Face

How to load it

from datasets import load_dataset

ds = load_dataset("risashinoda/AgroBench")