Rapidata/Imagen4_t2i_human_preference

这个数据集包含超过195k个人类响应，来自超过70k个个体标注者，仅在一天内收集。该数据集用于评估Imegen 4（imagen-4.0-ultra-generate-exp-05-20）在偏好、一致性和对齐性三个类别中的表现。评估包括Imegen 4与其他15个模型的1v1比较：Hidream I1 full、Halfmoon-4-4-2025、OpenAI 4o-26-3-25、Ideogram V2、Recraft V2、Lumina-15-2-25、Frames-23-1-25、Imegen-3、Flux-1.1-pro、Flux-1-pro、DALL-E 3、Midjourney-5.2、Stable Diffusion 3、Aurora和Janus-7b。

This dataset contains over 195k human responses from over 70k individual annotators, collected in just ~1 Day. It is used to evaluate Imegen 4 (imagen-4.0-ultra-generate-exp-05-20) across three categories: preference, coherence, and alignment. The dataset includes 1v1 comparisons between Imegen 4 and 15 other models such as Hidream I1 full, Halfmoon-4-4-2025, OpenAI 4o-26-3-25, Ideogram V2, Recraft V2, Lumina-15-2-25, Frames-23-1-25, Imegen-3, Flux-1.1-pro, Flux-1-pro, DALL-E 3, Midjourney-5.2, Stable Diffusion 3, Aurora, and Janus-7b.