mingiJ/origin_patch_inst_1-2_cot_prompt

该数据集是一个多模态问答数据集，包含258,268个训练样本。每个样本包含以下字段：唯一标识符（id）、提示文本（prompt）、提示类型（prompt_type）、问答类型（qa_type）、答案字母（answer_letter）、答案文本（answer_text）、数据来源（source）以及相关图像列表（images）。数据集支持文本和图像输入，适用于问答任务，如图像描述、选择题解答等。数据以train分割形式组织，总大小约224.72 MB。This dataset is a multimodal question-answering dataset containing 258,268 training examples. Each example includes the following fields: unique identifier (id), prompt text (prompt), prompt type (prompt_type), question-answering type (qa_type), answer letter (answer_letter), answer text (answer_text), data source (source), and a list of associated images (images). The dataset supports both text and image inputs and is suitable for QA tasks such as image captioning and multiple-choice question answering. It is organized into a single train split with a total size of approximately 224.72 MB.