Metraipolis › archive › dc_2953d28dcd39

Conservation Buys Stability and Factoring Buys Counterfactuals in Physical World Models — arXiv:2609.19674

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.19674
License (as archived)
CC0-1.0
Archived
2026-10-10T04:08:42.237Z
SHA-256 of the archived text
3f496cab7296639dbab231341f27ae586859ed7f88a75efdc8dcfe5583f09e93
Stable URL of this page
https://metraipolis.com/documento/dc_2953d28dcd39

How to cite

Conservation Buys Stability and Factoring Buys Counterfactuals in Physical World Models — arXiv:2609.19674 — arXiv (en), https://arxiv.org/abs/2609.19674 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2953d28dcd39 (Metraipolis archive dc_2953d28dcd39, sha256 3f496cab7296639dbab231341f27ae586859ed7f88a75efdc8dcfe5583f09e93)
BibTeX
@misc{dc2953d28dcd39,
  title = {Conservation Buys Stability and Factoring Buys Counterfactuals in Physical World Models — arXiv:2609.19674},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.19674},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2953d28dcd39; sha256 3f496cab7296639dbab231341f27ae586859ed7f88a75efdc8dcfe5583f09e93},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.19674. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

A learned simulator can reproduce its training conditions accurately yet fail in two distinct ways once those conditions change. Over long rollouts, small errors accumulate until the trajectory drifts away from physically plausible behavior; under an intervention on a physical parameter, the model may continue to follow the law seen during training rather than the intervened one. We show that these two failures require different structural remedies. Evolving a learned energy with a symplectic integrator preserves the geometry of the conservative dynamics and keeps rollouts bounded and physically meaningful for up to $100\times$ the training horizon, while equal-capacity predictors, an energy-regularized predictor, and a tuned neural ODE diverge. By contrast, encoding the physical coupling through an explicit linear factorization enables the model to follow a never-seen sign of that coupling, whereas an unrestricted parameterization remains locked to the training law. Crucially, the two mechanisms are separable: removing the structure responsible for long-horizon stability leaves counterfactual transfer intact, while removing the factorized coupling destroys counterfactual transfer without eliminating stability. This double dissociation, established with matched controls that remove or replace one structural component at a time, persists beyond the headline three-body system and remains visible when the physical state must be inferred from pixels rather than provided directly. The result is a concrete design principle for physical world models: long-horizon stability and changed-law generalization arise from distinct structural commitments, and each can be imposed deliberately without requiring the other.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-18 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.19674 — 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 2026-10-10T03:48:18.276Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).