kinredon/LEP

该数据集包含多个配置，每个配置都包含与图像和文本相关的多模态任务数据。具体配置包括ai2d_pruning、chartqa_pruning、cmmmu_val_pruning、docvqa_val_pruning、flickr30k_test_pruning、infovqa_val_pruning、mme、mmmu_val_pruning、nocaps_val_pruning、refcoco_bbox_val_pruning、seedbench_pruning、textcaps_val_pruning和textvqa_val_pruning。每个配置的特征包括问题、选项、答案、图像等字段，且每个配置都有不同的分割和示例数量。这些数据集主要用于图像和文本的多模态任务，如问答、图像描述等。

This dataset contains multiple configurations, each of which includes data related to multimodal tasks involving images and text. Specific configurations include ai2d_pruning, chartqa_pruning, cmmmu_val_pruning, docvqa_val_pruning, flickr30k_test_pruning, infovqa_val_pruning, mme, mmmu_val_pruning, nocaps_val_pruning, refcoco_bbox_val_pruning, seedbench_pruning, textcaps_val_pruning, and textvqa_val_pruning. Each configuration features fields such as questions, options, answers, images, etc., and each configuration has different splits and example counts. These datasets are primarily used for multimodal tasks involving images and text, such as question answering and image captioning.