Metraipolis › archive › dc_2dcffcec3673

H3-World: Turning Language Understanding into World Control — arXiv:2609.01560

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.01560
License (as archived)
CC0-1.0
Archived
2026-10-17T06:34:27.644Z
SHA-256 of the archived text
63251bef1beaff92797c515701e068cc0f90657e3aac783ef21d30f56c6b707f
Stable URL of this page
https://metraipolis.com/documento/dc_2dcffcec3673

How to cite

H3-World: Turning Language Understanding into World Control — arXiv:2609.01560 — arXiv (en), https://arxiv.org/abs/2609.01560 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2dcffcec3673 (Metraipolis archive dc_2dcffcec3673, sha256 63251bef1beaff92797c515701e068cc0f90657e3aac783ef21d30f56c6b707f)
BibTeX
@misc{dc2dcffcec3673,
  title = {H3-World: Turning Language Understanding into World Control — arXiv:2609.01560},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.01560},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2dcffcec3673; sha256 63251bef1beaff92797c515701e068cc0f90657e3aac783ef21d30f56c6b707f},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.01560. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We present H3-World, an efficient framework that turns the 33B MiniMax-H3 video generator into an interactive world model. Our key finding is that, as large video generators become more capable, language is emerging as a natural interface for control. MiniMax-H3, for example, already supports zero-shot control of character behavior and camera motion through natural-language instructions. Building on this, H3-World turns this coarse language interface into precise, temporally grounded world control, without introducing dedicated action modules. Specifically, we represent each action as a structured combination of character and camera instructions, and align them with the corresponding temporal video latents. To make the control temporally precise, we further introduce temporal attention routing, which restricts each instruction to its intended time interval and reduces control leakage across actions. Importantly, H3-World directly reuses the semantic representations learned during large-scale video pretraining and requires only lightweight adaptation. With only 8,000 gameplay samples, 10,000 LoRA optimization steps, and 0.199% trainable parameters, H3-World achieves effective character and camera control while preserving strong generation quality. It also generalizes to unseen scenarios. These results show that the control capabilities emerging in large video generators can be efficiently transformed into interactive world control.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-02 in cs.AI).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.01560 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-17T06:15:12.914Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).