SemVarBench

该数据集是一个评估在文本到图像合成中输入和输出之间语义变化因果关系的基准测试。它包含了一个名为SemVarEffect的新颖指标，用于评估在语义变化下，文本到图像合成模型的性能表现。该任务的目的是对文本到图像合成中的语义变化进行评估。This dataset is a benchmark for evaluating the causal relationship of semantic changes between inputs and outputs in text-to-image synthesis. It includes a novel metric named SemVarEffect to assess the performance of text-to-image synthesis models under semantic variations. The objective of this task is to evaluate semantic changes in text-to-image synthesis.