Epistemic Cultural Flattening in Generative Visual AI: Benchmarking Hungarian Heritage and Designing a V4 Path Toward Culturally Aware Text-to-Video

Name: Epistemic Cultural Flattening in Generative Visual AI: Benchmarking Hungarian Heritage and Designing a V4 Path Toward Culturally Aware Text-to-Video
Creator: Moholy-Nagy University of Art and Design
Published: 2026-05-25T01:00:45+08:00

Zenodo2026-05-25 更新2026-05-26 收录

数据链接：

https://zenodo.org/doi/10.5281/zenodo.20361254 数据链接链接失效反馈

官方服务：

资源简介：

Documentation package for 'Epistemic Cultural Flattening in Generative Visual AI: Benchmarking Hungarian Heritage and Designing a V4 Path Toward Culturally Aware Text-to-Video': prompt inventory, scoring-dimension definitions, codebook, and six aggregate summary tables for an expert-rated evaluation of four text-to-image models (Flux.1, SDXL, SD 3.5 Large, SD 3.5 Medium) across nine cultures and three domains (apparel, architecture, art).

提供机构：

Moholy-Nagy University of Art and Design

创建时间：

2026-05-25