Hunyuan-DiT：一种具有细粒度中文理解能力的强大多分辨率扩散 Transformer
计算机视觉与模式识别
2024-05-15 v1
摘要
我们提出了 Hunyuan-DiT，一种对英文和中文均有细粒度理解的文本到图像扩散 Transformer。为构建 Hunyuan-DiT，我们精心设计了 Transformer 结构、文本编码器和位置编码。我们还从零开始构建了完整的数据流水线，以更新和评估数据，从而迭代优化模型。为实现细粒度语言理解，我们训练了一个多模态大语言模型来精炼图像的描述文本。最后，Hunyuan-DiT 能够与用户进行多轮多模态对话，根据上下文生成并优化图像。通过我们包含 50 多名专业人类评估者的全面人工评估协议，与其他开源模型相比，Hunyuan-DiT 在中文到图像生成方面树立了新的 SOTA。代码和预训练模型已在 github.com/Tencent/HunyuanDiT 公开提供。
引用
@article{arxiv.2405.08748,
title = {Hunyuan-DiT: A Powerful Multi-Resolution Diffusion Transformer with Fine-Grained Chinese Understanding},
author = {Zhimin Li and Jianwei Zhang and Qin Lin and Jiangfeng Xiong and Yanxin Long and Xinchi Deng and Yingfang Zhang and Xingchao Liu and Minbin Huang and Zedong Xiao and Dayou Chen and Jiajun He and Jiahao Li and Wenyue Li and Chen Zhang and Rongwei Quan and Jianxiang Lu and Jiabin Huang and Xiaoyan Yuan and Xiaoxiao Zheng and Yixuan Li and Jihong Zhang and Chao Zhang and Meng Chen and Jie Liu and Zheng Fang and Weiyan Wang and Jinbao Xue and Yangyu Tao and Jianchen Zhu and Kai Liu and Sihuan Lin and Yifu Sun and Yun Li and Dongdong Wang and Mingtao Chen and Zhichao Hu and Xiao Xiao and Yan Chen and Yuhong Liu and Wei Liu and Di Wang and Yong Yang and Jie Jiang and Qinglin Lu},
journal= {arXiv preprint arXiv:2405.08748},
year = {2024}
}
备注
Project Page: https://dit.hunyuan.tencent.com/