wannabegosu/etri_mcq

该数据集未在README中直接描述，但根据task_categories和modalities字段，可以推断该数据集与视觉问答和视频分类任务相关，涉及视频和文本模态。configs部分列出了各种配置文件，表明数据集包含不同类型的动作和意图，如armflapping、headbanging和spinning，并有意图或无意图的变体。数据集为英文，大小在1K到10K条目之间。

The dataset is not directly described in the README. However, based on the task_categories and modalities fields, it can be inferred that the dataset is related to visual question answering and video classification tasks, involving video and text modalities. The configs section lists various configuration files, suggesting the dataset includes different types of actions and intents, such as armflapping, headbanging, and spinning, with variations indicating intent or no intent. The dataset is in English and falls under the size category of 1K to 10K entries.