Extended Data Fig. 5: Communication-qubit measurement and reset.

a, Pulse sequence for testing communication-qubit measurement and reset. The two communication qubits (transmons) are initialized in the joint state \((| gg\rangle +| ge\rangle +| eg\rangle +| ee\rangle )/2\). The two qubits are then measured and if the measurement indicates that the state is projected to \(| e\rangle \) a π-pulse is applied to flip the state to the ground state. Conditional quantum state tomography is performed to analyse the quality of measurement and reset. This measurement and reset protocol is used in the teleported gate. b, Experimentally measured Pauli vector components conditioned on the measurement outcome. We assign a ‘0’ (‘1’) to indicate that the measurement projected the qubit to be in \(| g\rangle \) \(\left(| e\rangle \right)\). For all outcomes, we find high fidelity to the two-qubit ground state \(| gg\rangle \), as expected, with ground-state fidelities of {00, 99.3%; 01, 95.7%; 10, 97.7%; 11, 94.2%}. From these results, we establish that the measurement and feedback processes for each qubit are independent; from the single-qubit reset infidelities, we expect a measurement fidelity of 1 − (0.993 − 0.957) − (0.993 − 0.977) = 0.948, which is consistent with the result for measurement outcome 11. c, Experimentally measured state after measurement-based reset. Measurement results from b are combined, and the compiled results illustrate that the reset protocol is high-fidelity and independent of the measurement outcome. The fidelity of this reconstructed two-qubit state to \(| gg\rangle \) is 96.9%.