arXiv:2609.32943
The wiring sets a phase-blind quantum memory's gap, the weight caps its coherence
Ruini Qian, Zhaobin Lyu, Zelong Yin, Jingjing Hu, Dengfeng Li, Shuoming An
Abstract
Engineered dissipation corrects errors, yet a memory's phase decays. What the memory's gap measures is not settled. Here we show that when each jump resets one basis state, the wiring can set the gap time while the phase keeps its rate. Under dephasing the gap time spans two orders across wirings; the phase does not. In this class, a contrast on two or more qubits fades faster than a bare qubit at any distance; no such pump power substitutes. The ceiling is two over the qubits the contrast spans. A two-branch pump can escape it at distance three; a two-clock test runs on published data.
AI Summary
The paper asks whether a dissipative quantum memory’s inverse Liouvillian gap measures its logical-coherence lifetime. It studies zero-Hamiltonian dynamics with independent dephasing, amplitude damping, and “phase-blind” reset jumps $L_\alpha=\sqrt{\kappa_\alpha}|\psi_\alpha\rangle\langle e_\alpha|$, whose inputs are individual computational-basis states. For opposite-phase superpositions of two basis codewords differing on $w$ qubits, it proves the all-time distinguishability bound $\|\Delta(t)\|_1\le e^{-wt/T_2}\|\Delta(0)\|_1$, where $T_2^{-1}=2\gamma_\phi+(2T_1)^{-1}$. Consequently, $T_X\le T_2/w$ and $T_X/T_1\le2/w$, independently of reset rates and assignments. The proof uses invariant fixed-XOR coherence sectors and a coefficient-norm contraction argument; parts are machine-checked in Lean 4. A four-qubit $ZZZZ$ plaquette illustrates the distinction between population and coherence modes. At $T_1=200\,\mu\mathrm{s}$, $T_\phi=100\,\mu\mathrm{s}$, and $\kappa=2\,\mathrm{MHz}$, enumerating all $8!$ bijective error-to-code assignments gives inverse gaps from $200$ to $40{,}300\,\mu\mathrm{s}$, while the GHZ coherence lifetime remains exactly $20\,\mu\mathrm{s}$. A closed classical population generator identifies the gap-setting mode. The bound concerns specified computational-basis contrasts, not arbitrary encoded states or recovery schemes. Non-diagonal control, superposition-valued reset inputs, and coherent multi-branch correction lie outside it. Numerical distance-three examples with coherent returns exceed bare-qubit coherence; the five-qubit code reaches $T_X/T_1=2.92$. The paper notes that the largest separations assume idealized dissipators, and that its coherent-correction advantages require gate fidelities above the demonstrated values considered.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.