About this paper
Jeveration repurposes a frozen vision-language model to generate coherent images and perform image-to-image and dense prediction tasks in a single forward pass with fast 512x512 generation.
System One models such as Jev return decisions through fast, parallel prediction. In this paper, we introduce Jeveration, showing that the same decision paradigm can also generate high-quality images in a single forward pass. Jeveration builds on a frozen vision-language model, repurposing its visual-semantic knowledge for holistic image generation. The key challenge is to maintain global consistency. A single prompt admits many valid images, yet parallel predictions must converge on a coherent whole. We therefore condition all positions on a shared random state to coordinate parallel predictions, so that each draw yields a different but coherent image. Since the model tends to ignore this…
System One modelsparallel predictionsingle forward passfrozen vision-language modelvisual-semantic knowledgeholistic image generationglobal consistencyshared random statedirect image supervisiondiscrete image codes
- Upvotes
- 4
- Authors
- 4
- Published
- Nov 18, 2026
Summary and abstract excerpt via Hugging Face Papers ↗ · arXiv.
Trending history
Jeveration: System One Models Now Generate was first seen on Hugging Face · Daily papers on Nov 19, 2026. Since then it has appeared in 12 checks on 2 different days. Its best position was #1 of 30, reached on Nov 19, 11:29 UTC. It dropped off the list after Nov 20, 03:29 UTC.