VLAI-AIVN/ViVQA-X

ViVQA-X是VQA-X数据集的越南语版本，专为越南语的视觉问答任务VQA-NLE设计。该数据集通过多阶段翻译和评估管道创建，旨在支持越南语视觉问答的研究，提供带有解释的翻译示例，与原始英文数据集保持一致。数据集包含问题和答案对，以及高质量的注释解释，图像来源于COCO数据集。

ViVQA-X is the Vietnamese version of the VQA-X dataset, designed for the Vietnamese language VQA-NLE task. This dataset is created through a multi-stage translation and evaluation pipeline, aiming to support research in Vietnamese visual question answering, providing translated examples with explanations that align with the original English dataset. The dataset includes question-answer pairs and high-quality annotated explanations, with images sourced from the COCO dataset.