QMS Platform: Representation Diagnostics and Real-Measurement Validation
Year: 2026
Type: software
Abstract: QMS Platform v0.4.0 introduces the provider-independent Three-Twin Architecture of the Quantum Measurement Stack (QMS): a finite two-mode Gaussian Quantum Twin for observability, state reconstruction, estimation, physicality checks, parameter identification and adaptation; a Detector Twin for residual analysis, stage-wise divergence, inverse calibration, mechanism ranking and detector-state assessment; and a Measurement Twin for causal prediction, innovation analysis, frozen calibration and experimental measurement-state monitoring. The release also separates provider-specific integrations from the QMS core and adds focused external-validation packages for the ESA-developed Pyxel detector-simulation framework and a Glasgow single-photon experimental dataset. The Pyxel pathway currently demonstrates controlled same-forward-model CTI parameter recovery and should be interpreted as an internal consistency test rather than independent detector validation. The Glasgow pathway demonstrates causal-replay measurement-state analysis on real experimental detector measurements with frozen calibration and future-only alerting. This release does not establish quantum-field reconstruction, independently validated real-detector calibration, physical fault attribution, universal detector health monitoring, statistically independent physical events, or prospective experimental validation. The current research focus is blind external validation using independently configured simulations and held-out experimental measurements.
Author: Srikar, R.
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7203973684