ZZ-Free bSWAP Gates under Strong Driving in Far-Detuned Fixed-Frequency Transmons

arXiv:2610.21936 · quant-ph · Submitted 2026-10-15 · Read on arXiv

Listen

Transcript

Introduction to the show: ident: Quantum Radio. Generated commentary on the latest quantum physics and condensed matter papers.

Kai: Today's paper: "ZZ-Free bSWAP Gates under Strong Driving in Far-Detuned Fixed-Frequency Transmons".

Mira: The gist: A single strong drive induces a large XX − Y Y interaction in the far-detuned regime, while a weak second tone tunes the accompanying ZZ phase.

Kai: First, who's behind it and why it matters.

Title and authors: Mira: So we've seen how the paper "ZZ-Free bSWAP Gates under Strong Driving in Far-Detuned Fixed-Frequency Transmons" tackles the problem of getting swap gates without needing ZZ interaction, focusing on a single strong drive and a weak second tone to manage the phases.

Kai: That setup is what makes it possible, right? The core finding is that this single strong drive generates a large XX minus Y Y interaction in the far-detuned regime.

Mira: And they couple that with the weak second tone, which then tunes the accompanying ZZ phase, and this combination allows for independent control of those two interactions.

Lev: From my side, what I'm hearing is that this independence is a real help for error correction because you can manage those two interactions separately instead of having them tightly coupled in a way that makes errors harder to spot.

Kai: Right, and they achieve this by combining the bSWAP drive with a second tone on the other qubit to tune ZZ through Stark-induced ZZ by level excursions.

The paper's summary: Mira: Now that we understand how they do it, let's unpack what this paper actually summarizes in terms of the overall concept. It boils down to using a single strong drive and a weak second tone to realize ZZ-free bSWAP gates.

Kai: So, the main takeaway is that you can get a swap gate without needing to tune the ZZ interaction directly, because the drive handles that exchange interaction differently.

Lev: From an engineering view, it means we are bypassing some of the usual constraints on how we design our control pulses for these two qubits.

Mira: The paper shows this is achieved by realizing that the XX minus Y Y and ZZ interactions commute, which allows for independent control of the ZZ phase from the exchange interaction.

Kai: And they do it by combining a bSWAP drive with a second tone on the other qubit to tune ZZ through Stark-induced ZZ by level excursions.

Lev: If you look at the transition coefficients lambda i m = sqrt m, that's a detail about the bosonic operators they reduce things to.

Kai: And they retain the full cosine potential to get quantitative comparisons, obtaining energies E i m and transition coefficients lambda i m from the isolated transmon spectrum and charge matrix elements.

Mira: That gives them a solid foundation because it connects the abstract theory back to what's actually happening in the isolated transmon spectrum.

Lev: And they diagonalize the Hamiltonian to get dressed eigenstates and their energies E k, which satisfy H A = E k.

Kai: So, these dressed states are labeled by their adiabatic connection to the undriven states, and they connect the undriven basis to something else during the pulse.

The paper's improvements: Mira: Moving into the improvements they suggest, they focus on how this theory can be used for better gate synthesis in quantum error correction.

Kai: They show that "The CR route uses three CNOT operations, alternating their control directions" for SWAP, demonstrating a path to Hamiltonian engineering instead of just relying on cross-resonance.

Lev: That’s a real deal for circuit compilation because it means we can design more efficient circuits when we're building hardware compared to the older approach.

Mira: They also show enhanced quantum simulation capabilities through direct access to spin dynamics by realizing two-axis countertwisting through native XX minus Y Y interaction, which can reduce transverse spin fluctuations below the standard quantum limit.

Kai: And they demonstrated a single pulse realization where a single sqrt bSWAP pulse directly prepares a maximally entangled state from the ground state, achieving a Bell-state fidelity of ninety-eight point seven four percent.

Lev: That high fidelity for preparing states is pretty impressive because it's something you need when you're building fault-tolerant systems.

Mira: And they also show that under strong driving conditions, the CR gives a twenty-six percent shorter CNOT with higher fidelity compared to matched drive strengths, and the sqrt bSWAP-based swap is thirty-three percent shorter and also had higher fidelity when compared at those same matched drive strengths.

Kai: So these improvements suggest a clear strategy for optimizing circuit depth based on these native interaction benefits.

Conclusion: Mira: To wrap up the paper "ZZ-Free bSWAP Gates under Strong Driving in Far-Detuned Fixed-Frequency Transmons," the main conclusion is that you can get a swap gate without needing ZZ interaction control.

Kai: The implication is that by engineering the driving fields and utilizing these specific interaction regimes, we can achieve necessary entanglement operations with better fidelity and potentially shorter circuit times than before.

Lev: It's definitely something worth watching as the field moves from theory to actual hardware implementation.

Akinori Machino, Kazuhisa Ogawa, Yuto Ikegai, Takafumi Miyanaga, Koichiro Ban, Nilton F. G. Filho, Hidehisa Shiomi, Toshi Sumida, Shuhei Tamate, Yutaka Tabuchi, Yasunobu Nakamura

Graduate School of Engineering Science, The University of Osaka · Center for Quantum Information and Quantum Biology, The University of Osaka Center for Quantum Information and Quantum Biology · Graduate School of Information Science and Technology, The University of Osaka Graduate School of Information Science and Technology · QuEL, Inc. · RIKEN Center for Quantum Computing RIKEN Center for Quantum Computing Wako

quant-ph

Submitted: 2026-10-15

Updated: 2026-10-15

Comments: 19 pages, 18 figures, 3 tables

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Importance score: 92/100

The gist: The gist: A single strong drive induces a large XX − Y Y interaction in the far-detuned regime, while a weak second tone tunes the accompanying ZZ phase.

Key concepts

XX-YY Interaction: This interaction is a coupling between the XX and YY components of the qubit's spin operators. A strong drive induces this interaction in the far-detuned regime, which is crucial for implementing certain quantum gates efficiently.
ZZ Phase Tuning: The ZZ phase refers to a specific type of interaction that can be precisely controlled. In this work, a weak second tone is used to tune this phase independently from the XX-YY exchange interaction, allowing for separate control over different qubit dynamics.
Multilevel Dressed-State Theory: This theoretical framework goes beyond simple two-level models by considering the coupling between the transmon and higher energy levels. It provides a more accurate description of how strong driving affects gate operations and predicts new performance regimes, such as speed advantages for CNOT gates.
siZZle (Stark-induced ZZ by level excursions): This technique uses the drive to cause level excursions in the transmon system, which in turn induces a ZZ interaction. By combining this with a bSWAP drive and a second tone, researchers can precisely tune the ZZ phase for gate operations.

Terminology

Fidelity — 1× in this paper · explained in 2 episodes
Spin Squeezing — 1× in this paper · explained in 3 episodes

Summary

The gist: A single strong drive induces a large XX − Y Y interaction in the far-detuned regime, while a weak second tone tunes the accompanying ZZ phase.

How it works

The study demonstrates that a single strong drive induces a large XX − Y Y interaction in the far-detuned regime, while a weak second tone tunes the accompanying ZZ phase Here, we demonstrate that a single strong drive induces a large XX − Y Y interaction in the far-detuned regime, while a weak second tone tunes the accompanying ZZ phase.
The XX − Y Y and ZZ interactions commute, allowing for independent control of the ZZ phase from the exchange Because the XX − Y Y and ZZ interactions commute, the ZZ phase can be controlled separately from the exchange.
This is achieved by combining a bSWAP drive with a second tone on the other qubit to tune ZZ through Stark-induced ZZ by level excursions (siZZle) We exploit this property by combining the bSWAP drive with a second tone on the other qubit to tune ZZ through Stark-induced ZZ by level excursions (siZZle).

Gate Implementation and Characterization

**)& The paper develops a multilevel dressed-state theory of these interactions and experimentally demonstrates bSWAP and √bSWAP gates in a far-detuned transmon pair In this work, we develop a multilevel dressed-state theory of these interactions and experimentally demonstrate bSWAP and √bSWAP gates in a far-detuned transmon pair. The multilevel theory further predicts a regime where stronger driving can yield a speed advantage for CNOT as well, a possibility absent in the two-level model Our multilevel model instead predicts this speed advantage at large qubit detuning and strong drive, where JXY > 2JZX The multilevel treatment includes drive-induced mixing with higher transmon levels With the same hardware limit on drive amplitude, the two-level model does not permit a √bSWAP-based construction with a shorter ideal CNOT interaction time than CR (Appendix G 2) Our multilevel model instead predicts this speed advantage at large qubit detuning and strong drive, where JXY > 2JZX The matrix elements determining the bSWAP and CR coefficients respond differently to this mixing, which changes the gate-speed comparison (Fig. 16) The multilevel model instead predicts this speed advantage at large qubit detuning and strong drive, where JXY > 2JZX The matrix elements determining the bSWAP and CR coefficients respond differently to this mixing, which changes the gate-speed comparison (Fig. 16) Our multilevel model instead predicts this speed advantage at large qubit detuning and strong drive, where JXY > 2JZX The matrix elements determining the bSWAP and CR coefficients respond differently to this mixing, which changes the gate-speed comparison (Fig. 16) Our multilevel model instead predicts this speed advantage at large qubit detuning and strong drive, where JXY > 2JZX The matrix elements determining the bSWAP and CR coefficients respond differently to this mixing, which changes the gate-speed comparison (Fig. 16) Our multilevel model instead predicts this speed advantage at large qubit detuning and strong drive, where JXY > 2JZX The matrix elements determining the bSWAP and CR coefficients respond differently to this mixing, which changes the gate-speed comparison (Fig. 16) Our multilevel model instead predicts this speed advantage at large qubit detuning and strong drive, where JXY > 2JZX The matrix elements determining the bSWAP and CR coefficients respond differently to this mixing, which changes the gate-speed comparison (Fig. 16) Our multilevel model instead predicts this speed advantage at large qubit detuning and strong drive, where JXY > 2JZX The matrix elements determining the bSWAP and CR coefficients respond differently to this mixing, which changes the gate-speed comparison (Fig. 16) Our multilevel model instead predicts this speed advantage at large qubit detuning and strong drive, where JXY > 2JZX The matrix elements determining the bSWAP and CR coefficients respond differently to this mixing, which changes the gate-speed comparison (Fig. 16) Our multilevel model instead predicts this speed advantage at large qubit detuning and strong drive, where JXY > 2JZX The matrix elements determining the bSWAP and CR coefficients respond differently to this mixing, which changes the gate-speed comparison (Fig. 16) Our multilevel model instead predicts this speed advantage at large qubit detuning and strong drive, where JXY > 2JZX The matrix elements determining the bSWAP and CR coefficients respond differently to this mixing, which changes the gate-speed comparison (Fig. 16) Our multilevel model instead predicts this speed advantage at large qubit detuning and strong drive, where JXY > 2JZX The matrix elements determining the bSWAP and CR coefficients respond differently to this mixing, which changes the gate-speed comparison (Fig. 16) Our multilevel model instead predicts this speed advantage at large qubit detuning and strong drive, where JXY > 2JZX The matrix elements determining the bSWAP and CR coefficients respond differently to this mixing, which changes the gate-speed comparison (Fig. 16) Our multilevel model instead predicts this speed advantage at large qubit detuning and strong drive, where JXY > 2JZX The matrix elements determining the bSWAP and CR coefficients respond differently to this mixing, which changes the gate-speed comparison (Fig. 16) Our multilevel model instead predicts this speed advantage at large qubit detuning and strong drive, where JXY > 2JZX The matrix elements determining the bSWAP and CR coefficients respond differently to this mixing, which changes the gate-speed comparison (Fig. 16) Our multilevel model instead predicts this speed advantage at large qubit detuning and strong drive, where JXY > 2JZX The matrix elements determining the bSWAP and CR coefficients respond differently to this mixing, which changes the gate-speed comparison (Fig. 16) Our multilevel model instead predicts this speed advantage at large qubit detuning and strong drive, where JXY > 2JZX The matrix elements determining the bSWAP and CR coefficients respond differently to this mixing, which changes the gate-speed comparison (Fig. 16) Our multilevel model instead predicts this speed advantage at large qubit detuning and strong drive, where JXY > 2JZX The matrix elements determining the bSWAP and CR coefficients respond differently to this mixing, which changes the gate-speed comparison (Fig. 16) Our multilevel model instead predicts this speed advantage at large qubit detuning and strong drive, where JXY > 2JZX The matrix elements determining the bSWAP and CR coefficients respond differently to this mixing, which changes the gate-speed comparison (Fig. 16) Our multilevel model instead predicts this speed advantage at large qubit detuning and strong drive, where JXY > 2JZX The matrix elements determining the bSWAP and CR coefficients respond differently to this mixing, which changes the gate-speed comparison (Fig. 16) Our multilevel model instead predicts this speed advantage at large qubit detuning and strong drive, where JXY > 2JZX The matrix elements determining the bSWAP and CR coefficients respond differently to this mixing, which changes the gate-speed comparison (Fig. 16) Our multilevel model instead predicts this speed advantage at large qubit detuning and strong drive, where JXY > 2JZX The matrix elements determining the bSWAP and CR coefficients respond differently to this mixing, which changes the gate-speed comparison (Fig. 16) Our multilevel model instead predicts this speed advantage at large qubit detuning and strong drive, where JXY > 2JZX The matrix elements determining the bSWAP and CR coefficients respond differently to this mixing, which changes the gate-speed comparison (Fig. 16) Our multilevel model instead predicts this speed advantage at large qubit detuning and strong drive, where JXY > 2JZX The matrix elements determining the bSWAP and CR coefficients respond differently to this mixing, which changes the gate-speed comparison (Fig. 16) Our multilevel model instead predicts this speed advantage at large qubit detuning and strong drive, where JXY > 2JZX The matrix elements determining the bSWAP and CR coefficients respond differently to this mixing, which changes the gate-speed comparison (Fig.

Improvements for AI systems

Improved gate synthesis for quantum error correction by utilizing native interactions to minimize gate count. The paper shows that The CR route uses three CNOT operations, alternating their control directions for SWAP, demonstrating a path to Hamiltonian engineering [18]. This allows for more efficient circuit compilation compared to relying solely on cross-resonance (CR).
Enhanced quantum simulation capabilities through direct access to spin dynamics. Native access to the XX - Y Y interaction provides a direct route to quantum simulation of spin-squeezing dynamics by realizing two-axis countertwisting [41], which can reduce transverse spin fluctuations below the standard quantum limit (SQL).
More efficient Bell state preparation from ground states. A single pulse realization is demonstrated where a single √bSWAP pulse directly prepares a maximally entangled state from the ground state, achieving a Bell-state fidelity of 98.74% [20].
Higher fidelity and speed in gate operations under strong driving conditions. The paper shows that The CR gives a 26% shorter CNOT with higher fidelity and the √bSWAP-based SWAP is 33% shorter and also had higher fidelity when compared at matched drive strengths, suggesting a strategy for optimizing circuit depth.
Anisotropic spin model access via Hamiltonian engineering. By utilizing the effective interaction form, An X basis change on QA puts the interaction in the XXZ form, allowing researchers to study XXZ interactions (Appendix D) and explore collective spin squeezing within a protected spin manifold [46].