cd /news/artificial-intelligence/v-rae-rethinking-video-latent-spaces… · home topics artificial-intelligence article
[ARTICLE · art-97016] src=arxiv.org ↗ pub= topic=artificial-intelligence verified=true sentiment=· neutral

V-RAE: Rethinking Video Latent Spaces for Generation

Researchers propose V-RAE, a video representation autoencoder that builds compact generative latents on top of frozen vision foundation model representations, achieving 2.13 rFVD on K600 and outperforming all evaluated large-scale pretrained video VAEs. Under matched generation settings, the best variant achieves gFVD scores of 117.86 and 19.16 on UCF101 and K600, respectively, while converging up to 6x faster. The study introduces tFVD, a temporal-coherence diagnostic that correlates more reliably with downstream generation quality, and shows that frozen semantic representations can support video reconstruction, generation, and predictive modeling.

read2 min views1 publishedAug 14, 2026
V-RAE: Rethinking Video Latent Spaces for Generation
Image: source
[Submitted on 13 Aug 2026]


[View PDF](/pdf/2608.13556)

[HTML (experimental)](https://arxiv.org/html/2608.13556v1)

Abstract:Latent video generation relies on autoencoders to define a compact space in which generative models operate. Although video autoencoder architectures have evolved substantially, their latent spaces are still optimized primarily for pixel-level reconstruction and provide limited high-level semantic organization. A reconstruction-optimal latent space, however, need not be well suited to generative modeling. We propose V-RAE, a video representation autoencoder that builds compact generative latents on top of frozen vision foundation model representations. A lightweight temporal pooling module removes temporal redundancy while preserving semantic structure, and a video decoder reconstructs continuous motion from the compressed features. We evaluate V-RAE with four representative frozen encoders on video reconstruction, semantic probing, and class-conditional generation. V-RAE achieves 2.13 rFVD on K600, outperforming all evaluated large-scale pretrained video VAEs. Its latents retain substantially more semantic information than conventional video tokenizer latents. Under matched generation settings, our best variant achieves gFVD scores of 117.86 and 19.16 on UCF101 and K600, respectively, while converging up to 6x faster}. We further show that reconstruction quality alone is insufficient to characterize generative utility and introduce tFVD, a temporal-coherence diagnostic that correlates more reliably with downstream generation quality. Beyond video generation, V-RAE also improves future video prediction on Cityscapes over the Wan 2.2 VAE latent space under matched prediction settings. Taken together, the experiments show that frozen semantic representations can support video reconstruction, generation, and predictive modeling. The project page:[this https URL].

References & Citations

...

Bibliographic Explorer

(What is the Explorer?) Connected Papers

(What is Connected Papers?) Litmaps

(What is Litmaps?) scite Smart Citations

(What are Smart Citations?)# Code, Data and Media Associated with this Article alphaXiv

(What is alphaXiv?) CatalyzeX Code Finder for Papers

(What is CatalyzeX?) DagsHub

(What is DagsHub?) Gotit.pub

(What is GotitPub?) Hugging Face

(What is Huggingface?) ScienceCast

(What is ScienceCast?)# Demos Influence Flower

(What are Influence Flowers?) CORE Recommender

(What is CORE?)# arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

── more in #artificial-intelligence 4 stories · sorted by recency
── more on @v-rae 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/v-rae-rethinking-vid…] indexed:0 read:2min 2026-08-14 ·