krutrim-ai-labs/IndicMMVet

IndicMMVet是一个包含12种印度语言及英语的视觉问答数据集，共218个样本。每个样本包含一个问题、对应的能力类型（如识别、OCR、知识、语言生成、空间意识和数学）以及答案。该数据集用于评估大型视觉语言模型在集成多种核心能力方面的表现。

IndicMMVet is a Visual Question Answering (VQA) dataset containing 218 samples across 12 Indic languages and English. Each sample includes a question, a type of capability (such as recognition, OCR, knowledge, language generation, spatial awareness, and math), and an answer. The dataset is designed to evaluate the performance of large vision-language models in integrating multiple core capabilities.