# DreamX preprint reports 300× compression of video action representations

_Published Wednesday, November 4, 2026 at 1:06 AM EST · AI, Robotics, Science · Latest · Tier 2 — Notable_

Researchers report in a DreamX preprint that they compressed each latent video frame from 1,200 tokens to 4 while retaining action information that transfers across human and robot embodiments and environments.

The method repurposes a pretrained video diffusion model, extracting compact representations of actions from features that capture movement over space and time. The authors report leading performance among latent action models on LARY-Bench and action transfer without additional training across different embodiments and environments.

The shared representation also supports policy learning in a compact action space, reducing the computational cost of video-diffusion-based policies. The findings are reported in a preprint.

## Sources

- [arXiv Query: search_query=cat:cs.RO&id_list=&start=0&max_results=30](https://arxiv.org/abs/2611.05377v1)

---
Canonical: https://techandbusiness.org/newswire/GUJU_EjZBmmmvYJP1PrGUd
Published: 2026-11-04T06:06:09.673Z
Story chronology: 2026-11-03T18:59:56.000Z
Retrieved: 2026-11-04T09:10:14.039Z
Publisher: Tech & Business (techandbusiness.org)