ThinkMorph/Spatial_Navigation

这是一个为训练ThinkMorph-7B模型而创建的增强型交织数据集，专注于四种代表性的任务：拼图组装、空间导航、视觉搜索和图表重聚焦，这些任务需要不同程度的视觉参与和跨模态交互。数据集包含问题ID、问题文本、正确答案、问题图像、推理思路文本段、推理中间图像、第二个推理文本段、任务类型和纯文本推理轨迹。

This is an enriched interleaved dataset created for training the ThinkMorph-7B model, focused on four representative tasks: Jigsaw Assembly, Spatial Navigation, Visual Search, and Chart Refocus, which require varying degrees of visual engagement and cross-modal interactions. The dataset includes problem ID, question text, correct answer, problem image, reasoning thought text segment, reasoning intermediate image, second reasoning text segment, task type, and text-only reasoning trajectory.