THE QUANTUM INTELLIGENCER

A paper deposited on the electronic archives proposes a new mechanism for controlling the interaction between two superconducting qubits: a double-resonator coupler, in which two resonators are joined by a single Josephson junction and a capacitor. In simulation, the hybridised resonator modes create two exchange paths whose interference can suppress the residual ZZ interaction entirely when the qubits sit outside the usual straddling frequency range, even with no direct qubit-qubit coupling; when driven away from that idle point, the same circuit yields a ZZ interaction of roughly seventy megahertz, enough for a twenty-nanosecond controlled-Z gate with coherent infidelity calculated below one part in one hundred thousand. That is the design as proposed. No hardware has yet been built to test it, and the fidelity figures remain products of simulation rather than measurement. A preprint is a proposal awaiting the bench; this one, nonetheless, is a proposal of unusual flexibility in frequency allocation and qubit placement, and prudent engineers will wish to catalogue it. The paper opens by marking where tunable couplers now stand: two-qubit gate fidelities approaching 99.9 per cent, with the simultaneous suppression of residual interactions and the flexible allocation of qubit frequencies still unresolved. The abstract names its figure a simulated coherent infidelity. The adjective is doing exact work: it tells the reader that nothing in the calculation decays, drifts, or dephases. A coherent infidelity below one part in one hundred thousand is therefore a statement about the idealised excursion of the state, not about the error a physical device would exhibit. Likewise, the complete cancellation of the residual ZZ interaction is a property of the model's chosen operating point; the paper does not show that a manufactured coupler can be tuned to that point and kept there despite fabrication spread and control noise. The proposal, as presented, concerns a single coupler between two qubits. Its claims of residual-interaction cancellation and a twenty-nanosecond gate are made for that pair. No account is given in the abstract of how such couplers behave in a larger array, what calibration they exact, or what crosstalk they introduce. Those are separate questions, and they require separate work. A focused result earns its place as a component; it is not a blueprint for the whole machine. —Ada H. Pemberley Dispatch from The Prepared E0