Author

Minghui Guo

h-index 8 View on Semantic Scholar

We have 1 of 30 papers

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

Preprint Aug 2026

V-RAE: Rethinking Video Latent Spaces for Generation

V-RAE, a video representation autoencoder that builds compact generative latents on top of frozen vision foundation model representations, and tFVD, a temporal-coherence diagnostic that correlates more reliably with downstream generation quality are introduced.

Ming-Hui Guo, Sheng-Qiong Wu, Hao Fei · 3 citations