Quantum Measurement Stack (QMS): A Three-Twin Architecture for Measurement Observability and Adaptive Inference

The Quantum Measurement Stack (QMS) is an open-source research and software framework for studying what can be inferred about a physical or quantum system from what its detector and measurement architecture make observable. This technical note documents the QMS v0.4.0 Three-Twin Architecture comprising a finite two-mode Gaussian Quantum Twin for observability, reconstruction, estimation and adaptation; a Detector Twin for residual analysis, stage-wise divergence, inverse calibration and mechanism ranking; and a Measurement Twin for causal prediction, innovation analysis, frozen calibration and experimental measurement-state monitoring. Current evidence includes controlled computational validation, same-forward-model CTI parameter recovery using the Pyxel detector-simulation framework, and causal-replay analysis of archival University of Glasgow single-photon experimental measurements. The current evidence does not establish quantum-field reconstruction, general quantum-state tomography, independently validated real-detector calibration, detector-health diagnosis, physical fault attribution, statistically independent physical-event discovery, or prospective experimental validation. The next major milestone is blind external validation using independently configured simulations and held-out experimental measurements.