allenai/Molmo2-MultiImageQA

Molmo2-MultiImageQA是PixMo-AskModelAnything数据集的**多图像扩展**版本，这是一个用于视觉语言模型指令调优的数据集。它包含人类编写的**多图像**问答对，具有长形式的答案。该数据集是[Molmo2数据集集合](https://huggingface.co/collections/allenai/molmo2-data)的一部分，并用于训练[Molmo2模型家族](https://huggingface.co/collections/allenai/molmo2)。每个示例包含一个有序的图像URL列表和多个基于这些图像的相关问答对。

Molmo2-MultiImageQA is a **multi-image extension** of [PixMo-AskModelAnything](https://huggingface.co/datasets/allenai/pixmo-ask-model-anything), an instruction-tuning dataset for vision-language models. It contains human-authored question-answer pairs over **multiple images** with long-form answers. Molmo2-MultiImageQA is part of the [Molmo2 dataset collection](https://huggingface.co/collections/allenai/molmo2-data) and was used to train the [Molmo2 family of models](https://huggingface.co/collections/allenai/molmo2). Each example contains an ordered list of image URLs and multiple related question–answer pairs grounded in those images.