S1b IBM Direct-XY Observable Result

Date: 2026-05-20

Backend: ibm_kingston

Parent S1 result: docs/campaigns/s1_ibm_feedback_pair_result_2026-05-20.md

Purpose

S1b is an extension of the same S1 paper, not a separate paper. The first S1 hardware run showed that the binary synchrony proxy was near saturation in both arms. S1b keeps the same dynamic-feedback and matched open-loop bodies but measures direct XY-sector Pauli correlators at final readout.

Jobs

Observable	Feedback job	Matched open-loop job
XXI	d86r1rqs46sc73f7c2g0	d86r252s46sc73f7c2tg
YYI	d86r1udg7okc73elggi0	d86r26gp0eas73dlbkkg
IXX	d86r201789is739022q0	d86r288p0eas73dlbkmg
IYY	d86r21is46sc73f7c2o0	d86r29p789is7390238g

Artefacts:

• data/s1_feedback_loop/s1b_xy_observable_readiness_ibm_kingston_20260520T130238Z.json
• data/s1_feedback_loop/s1b_xy_observable_raw_counts_ibm_kingston_20260520T130238Z.json
• data/s1_feedback_loop/s1b_xy_observable_analysis_ibm_kingston_20260520T130238Z.json

Result

Observable	Feedback mean	Matched open-loop mean	Feedback minus control
IXX	0.9042968750	0.8795572917	0.0247395833
IYY	0.8977864583	0.8619791667	0.0358072917
XXI	0.8424479167	0.8600260417	-0.0175781250
YYI	0.8632812500	0.8593750000	0.0039062500

Mean absolute feedback-control separation across the four direct XY channels: 0.0205078125.

Interpretation

S1b removes the saturated binary-synchrony proxy and shows that the feedback/control difference is small but channel-structured. Three of four direct XY channels move positive for the feedback arm, while XXI moves negative. This supports a conservative paper claim: the tested feedback policy does not deliver robust target control, but direct XY-sector measurements reveal a non-uniform feedback/control response hidden by the binary proxy.

Boundary

This does not establish backend-general feedback control, quantum advantage, or sub-microsecond real-time control. It is a bounded dynamic-circuit hardware-control stress test on the selected backend, circuit family, shots, repetitions, and calibration window.