task_categories: - visual-question-answering - question-answering language: - en - vi size_categories: - 100K<n<1M tags: - multi-modal - image-captioning - vqa - sharegpt4v - GPT4-Vision