Metraipolis › archive › dc_cb40e07bbdaa

Observable-Specific Validity Tests for Scientific AI: Controlled Approximations and Intervention Response — arXiv:2610.41939

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.41939
License (as archived): CC0-1.0
Archived: 2026-11-03T05:33:37.989Z
SHA-256 of the archived text: bcd86e16fc298b24e288dd070c82e67bd486952601f716266908e27456a9ca25
Stable URL of this page: https://metraipolis.com/documento/dc_cb40e07bbdaa

How to cite

Observable-Specific Validity Tests for Scientific AI: Controlled Approximations and Intervention Response — arXiv:2610.41939 — arXiv (en), https://arxiv.org/abs/2610.41939 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_cb40e07bbdaa (Metraipolis archive dc_cb40e07bbdaa, sha256 bcd86e16fc298b24e288dd070c82e67bd486952601f716266908e27456a9ca25)

BibTeX

@misc{dccb40e07bbdaa,
  title = {Observable-Specific Validity Tests for Scientific AI: Controlled Approximations and Intervention Response — arXiv:2610.41939},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.41939},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_cb40e07bbdaa; sha256 bcd86e16fc298b24e288dd070c82e67bd486952601f716266908e27456a9ca25},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.41939. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

An approximation can reproduce an energy while missing its response, or satisfy an error tolerance without being certified by a given bound. We formulate observable-specific validity records for scientific artificial intelligence (AI) that link a reduction to its reference model, error measure, tolerance, parameter domain, and transformations or interventions. Explicit displacement and velocity certificates for a singularly perturbed oscillator recover distinct convergence orders and provide sufficient validity guarantees. Exact Hubbard dimer relations resolve the different coupling domains accepted by energy and response criteria. Across 16 repeated neural fits, imposing dimensional homogeneity reduces energy scale transfer error from 52.3% to 0.0768% and raises the median fraction jointly satisfying gap and response tolerances from zero to 100%. Paired transformations verify the protected scaling relation, while a coupling crossover identifies the additional information needed for response transfer. A causal model recovers a positive intervention response beneath a strong negative observational association. For two mechanisms observed with Gaussian noise whose responses first differ at order $x^k$, we derive an exact fixed control query requirement scaling as $x^{-2k}$ and verify its full success curve numerically. Together with a 256-task evaluation interface, these results provide independently checkable tests of controlled approximation, physical transfer, and mechanism discrimination in scientific AI.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI); Digital Libraries (cs.DL).
Pagina del lavoro: https://arxiv.org/abs/2610.41939 — 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 2026-11-03T05:18:52.234Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).