# Semantic World Models: Using Vision-Language Models to Predict Semantic Information for Robot Planning

- Source: arXiv cs.RO
- Published: 2026-10-28T04:00:00.000Z
- RoboHot: https://robohot.news/en/items/zbcjrxm4zmos5vns3471j06y1
- Original: https://arxiv.org/abs/2510.19818

## Summary

The paper proposes Semantic World Models, which model the world model as visual question answering about the semantic information of future frames rather than reconstructing future pixels, so that vision-language models can be obtained through supervised fine-tuning on image-action-text data. The authors validated it on simulated robot tasks, where it outperformed typical reconstruction-based action-conditioned world models.