Metraipolis › archive › dc_f033cce3e675
Flex-$π$: A Multi-Stream World-Action Model with Compute Flexibility — arXiv:2608.10860
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.10860
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-19T06:30:49.325Z
- SHA-256 of the archived text
816e8b5742abfecaa31d5d2fa1380f485854de7e0298b6091e3f44736352d5d4- Stable URL of this page
https://metraipolis.com/documento/dc_f033cce3e675
How to cite
Flex-$π$: A Multi-Stream World-Action Model with Compute Flexibility — arXiv:2608.10860 — arXiv (en), https://arxiv.org/abs/2608.10860 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f033cce3e675 (Metraipolis archive dc_f033cce3e675, sha256 816e8b5742abfecaa31d5d2fa1380f485854de7e0298b6091e3f44736352d5d4)
BibTeX
@misc{dcf033cce3e675,
title = {Flex-$π$: A Multi-Stream World-Action Model with Compute Flexibility — arXiv:2608.10860},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.10860},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f033cce3e675; sha256 816e8b5742abfecaa31d5d2fa1380f485854de7e0298b6091e3f44736352d5d4},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.10860. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
World-action models (WAMs) predict the future to act better, but nearly all of them predict only RGB latents, trained purely for pixel reconstruction, with no explicit signal for the 3D geometry or object semantics manipulation needs. We find a surprising free lunch: the same frozen video-generation VAE that encodes RGB also encodes 3D pointmaps almost losslessly, with no pointmap-specific training at all. This lets us supervise Flex-$\pi$, a 6B-parameter WAM, on 3D geometry and object-centric DINO semantics alongside RGB, at no cost in new sensors, new pre-training, or inference latency. Every visual signal is projected into this shared latent space and denoised jointly with actions inside a Mixture-of-Transformers backbone; per-stream dropout with cross-modality forcing then lets a single trained checkpoint run on any subset of these streams, from a fast action-only mode to full joint generation. The result is a policy that is exceptionally demonstration-efficient and generalizes well, beating the strongest baselines by up to 2-7$\times$ on dexterous, precise, real-world bimanual manipulation tasks both in and out of distribution, all while running faster than $\pi_{0.5}$. Our project website: this https URL
Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-12 in cs.CV).
Categorie arXiv: Robotics (cs.RO) ; Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2608.10860 — 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 2026-10-19T06:25:27.348Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-08-12?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).