5CD-AI/Viet-LLaVA-CoT-o1-Instruct

该数据集包含了图像和与图像相关的文本信息，包括问题（question）和输出（output），以及它们的越南语版本（vi_question和vi_output）。每个样本还有一个唯一的标识符（id），以及可能的额外标识（id_num）。此外，还有一个地面真实字段（ground_truth），可能是用于标注或验证的。数据集分为训练集（train），共有53167个样本。

The dataset includes images and associated text information, such as questions (question) and outputs (output), as well as their Vietnamese versions (vi_question and vi_output). Each sample also has a unique identifier (id) and possibly an additional identifier (id_num). Moreover, there is a ground truth field, which might be used for annotation or validation. The dataset is split into a training set (train) with a total of 53,167 samples.