Year: 2026
Type: preprint
Abstract: An unreliable language model can be made to produce reliable physical designs if the authority to assert is moved out of the model: the model proposes, and a deterministic engine alone certifies, returning certified, impossible, or unknown. We introduce Physics-Anchored Certification (PHACT), a propose-certify loop spanning five scientific domains, and identify what makes such a certificate trustworthy. A checker that accepts a model-supplied value can be forged; deriving the certified quantity from fixed inputs instead makes forgery impossible by construction. Across eighty adversarial trials spanning two models, two decoding temperatures, and a deliberately faulted engine, this contract produced zero false certifications.
Source: arXiv (Cornell University)
Authors: Vyas, Nakul, Stoev, Iliya D.
Institution: Karlsruhe Institute of Technology
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
Funders: Nvidia, Universität Stuttgart
OpenAlex ID: W7166665959
Cites:
Cited by:
Related to: