maum-ai/KOFFVQA_Data
收藏官方服务：
资源简介：
KOFFVQA是一个韩语通用视觉语言模型基准，专注于自由形式的视觉问题回答。该数据集包含图像、与图像相关的问题以及用于评估响应的评分标准，旨在评估大型视觉语言模型对图像相关问题生成全面准确文本回答的能力。
KOFFVQA is a general-purpose VLM benchmark in the Korean language, focusing on free-form visual question answering. The dataset includes images, corresponding questions, and response grading criteria, aiming to evaluate the ability of large vision-language models to generate comprehensive and accurate text responses to questions about images.
提供机构：
maum-ai
