BLNL/AmbiBench
收藏官方服务：
资源简介：
这个数据集提供了用于AmbiBench基准测试的元数据，包含模糊的图像或视频样本及其对应的问答对。适用于训练和评估视觉-语言模型在模糊图像理解方面的性能。
This dataset provides metadata used in the AmbiBench benchmark, including ambiguous image or video samples and their corresponding question-answer pairs. It is suitable for training and evaluating vision-language models on ambiguous image understanding.
提供机构：
BLNL
