latent-compression

GRACE提出两阶段生成感知潜空间压缩框架，在保持与预训练DiT兼容的前提下将视频自编码器压缩率提升，使Wan2.1-I2V-14B的token数减少8倍、延迟降低11.1倍，同时在VBench上保持同等生成质量。

This paper introduces generic triple-latent recurrent models that compress token pair interactions into a latent state, and a gated associative retrieval variant that improves exact recall. The hybrid model outperforms Transformers on byte-level WikiText-2 and a tokenized language benchmark, achieving up to 41.9% associative recall versus 25%.