Hy-Nomas: Towards Free-form Interactive World Generation

Hy-Nomas, a video world model built on an 8B diffusion backbone, was presented as a system for free-form interactive world generation that follows successive text commands and continuous camera control while preserving relevant history and action consequences across multiple interactions. The model is designed to let users issue free-form text instructions alongside camera movement rather than rely on fixed command sets.

Interactive world generation requires models to follow successive commands while preserving relevant history and action consequences. We present Hy-Nomas, a video world model with an 8B diffusion backbone that supports free-form text instructions and continuous camera control across multiple interac