# ALPS: Visual State Encoders for World Models

- Source: HuggingFace Daily Papers
- Published: 2026-10-21T00:00:00.000Z
- RoboHot: https://robohot.news/en/items/h5a6xaxs66thz3ngrd53kyjzp
- Original: https://huggingface.co/papers/2610.30120

## Summary

The study proposes ALPS, a pretrained visual state encoder that unifies three capabilities—physical state information, language alignment, and pixel reconstruction—within a single set of patch tokens. It pretrains a ViT-B/16 from scratch using only still images and text, and further refines it through cross-view self-distillation.