什么是 DiT？

ByteDance open-sources Bernini, a unified framework for video generation and editing that uses a multimodal large model (MLLM) to understand semantic instructions first, then delegates high-quality rendering to a DiT diffusion model, enabling a paradigm shift from 'listening to prompts' to 'understanding before acting' in AI video creation, supporting controllable editing and reference-based generation.

苹果提出基于解耦时间深度扩散变换器的音频合成架构，实现21MB内存下10ms/步的实时音频生成，MOS评分提升0.28。

The interpretation of the DiT paper is very difficult, despite a detailed 13,000-word explanation, there are still many parts that are hard to understand.