Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Structural Certification for Reliable Physical Design with Language Models

About

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.

Nakul Vyas, Iliya D. Stoev• 2026

Related benchmarks

TaskDatasetResultRank
Certified Task-solvingPHACT Safety Suite Feasible goals--
2
Certified Task-solvingPHACT Safety Suite (Impossible goals)--
2
Feasible CertificationDRONE
Certifications Passed8
1
Feasible CertificationBioChem
Certification Success Rate0.8
1
Feasible CertificationAstrophysics
Certified Success Score9
1
Feasible CertificationElectrical
Certified100
1
Feasible CertificationDrug formulation
Certified Success Rate0.9
1
Feasible CertificationPHACT Evaluation Suite Overall
Certified Rate44
1
Showing 8 of 8 rows

Other info

Follow for update