Pauli Cloners: Programmable Quantum Cloning

• Pauli Cloners are quantum cloning machines engineered to replicate outputs as Pauli channels using programmable ancilla states and structured gate circuits.
• They support various cloning strategies—including universal, phase-covariant, and biased approaches—by optimizing fidelity metrics for quantum communication tasks.
• Applications span programmable attacks on QKD, error mitigation in noisy channels, and efficient broadcasting of entanglement and quantum correlations.

A Pauli cloner is a quantum cloning machine whose action on each output is specifically engineered to resemble a Pauli channel—the class of quantum channels characterized by convex mixtures of Pauli operators acting on $N$-qubit systems. The fundamental feature of a Pauli cloner is its programmable disturbance, which can be tailored such that the output states incur only Pauli-type noise, and therefore faithfully realize a wide variety of optimal and biased cloning transformations in both single- and multi-qubit scenarios. Such cloners provide the architecture for programmable attacks on quantum communication protocols, enable fine-grained control in quantum error-mitigation, and underpin analyses of broadcasting entanglement and quantum correlations subject to practical noise constraints (Kerstan et al., 31 Jan 2026, Jain et al., 2017).

1. Structural Definition and Circuit Architecture

A Pauli cloner is formally defined as a one-to-two quantum cloning apparatus represented by a global unitary $U$ acting on an input register, associated ancilla, and yielding three subsystems: two clones (conventionally denoted $A$ and $B$) and an environment $E$ ("Eve"). Upon tracing out $E$, each clone is subject to a completely positive trace-preserving (CPTP) map of strictly Pauli form: $$\mathcal{E}(\rho) = \sum_{P\in\mathcal P_N} p_P\, P \rho P,$$ where $\mathcal P_N = \{I,X,Y,Z\}^{\otimes N}$ is the $N$-qubit Pauli group and $\{p_P\}$ a probability distribution. The correspondence of the cloned outputs to Pauli channels distinguishes Pauli cloners from more general, possibly non-Pauli stochastic cloners. The circuit instantiation is an extension of the Niu–Griffiths (NG) design: the hardware comprises staggered layers of CNOT gates and Hadamards, while the software is encoded in the initial (program) state of the ancilla registers. For $N$ qubits, the circuit acts on $3N$ physical qubits, with gate networks and ancilla-state preparation scaling linearly in $N$ (Kerstan et al., 31 Jan 2026).

2. Explicit Construction: Single- and Multi-Qubit Pauli Cloners

In the single-qubit case, the programmable ancilla state is

$$|\psi\rangle = (a, b, c, d)^T, \qquad a^2 + b^2 + c^2 + d^2 = 1.$$

The corresponding fidelity in basis $Z$, $X$, and $Y$ for Bob’s clone ($F_{AB}$) is set by squared amplitudes, e.g. $F_{AB,Z}=a^2+c^2$, while Eve’s fidelities are functions of both the amplitudes and their relative phases. Optimization of these parameters yields several notable instances:

• Asymmetric universal Pauli cloner (UQCM): Constraints $F_{AB,X} = F_{AB,Y} = F_{AB,Z} = F$ select a one-parameter family generalizing Cerf's asymmetric UQCM, with optimal single-copy fidelities and explicit ancilla coefficients.
• Phase-covariant Pauli cloner (PCCM): Setting $F_{AB,Z}=1/2$ and optimizing $F_{AB,X}=F_{AB,Y}$ yields the phase-covariant cloner, particularly relevant for BB84-type QKD attacks where only certain bases are targeted for high-fidelity replication.
• Biased/imbalanced Pauli cloner: Where the channel is dominated by particular Pauli noise, the software parameters can be optimized to privilege fidelity in less affected bases, resulting in closed-form expressions for ancilla populations that maximize a weighted fidelity functional.

The $N$-qubit generalization employs $(\mathbb{C}^2 )^{\otimes 2N}$ ancilla registers with amplitudes $\{a_P\}$, yielding clone fidelities in each MUB $M_i$ as $F_{AB, M_i} = \sum_{P \in G_i} a_P^2$, with $G_i$ the maximal commuting subgroup of Pauli operators fixing $M_i$ (Kerstan et al., 31 Jan 2026).

3. Pauli Errors, Mutually Unbiased Bases, and Cloner Optimality

There is a fundamental connection between the structure of Pauli errors, the algebraic decomposition of the Pauli group into maximal commuting subgroups, and the resulting performance of Pauli cloners on states in different mutually unbiased bases (MUBs). Pauli operators act nontrivially within all but one basis (which they fix, up to phase), so the design of a given Pauli cloner can be efficiently matched to quantum tasks encoded in various MUBs. For universal symmetric cloning ($a_P^2=1/4^N$ for all $P$), the fidelity on any MUB is

$F = \frac{2^N+3}{2(2^N+1)},$

recovering the optimal symmetric UQCM performance (Kerstan et al., 31 Jan 2026).

4. Tailoring to Noise Models and Quantum Communication

By programming the ancilla state, Pauli cloners can be dynamically optimized for arbitrary Pauli channel parameters $\{p_P\}$ representing observed or expected noise regimes. In quantum key distribution (QKD), this enables an adversary to tailor a clone attack to complement channel-induced errors, thereby maximizing cloning fidelity under channel tomography feedback. Conversely, legitimate users may inject specifically structured Pauli noise to counteract or frustrate a Pauli-cloner attack—a practical error-mitigation application (Kerstan et al., 31 Jan 2026).

In a noisy Pauli channel $\mathcal E$, the effective realized fidelity on a basis $M_i$ is

$$\tilde F_{AB,M_i} = F_{AB,M_i}(1 - 2 \sum_{Q \notin G_i} p_Q) + \sum_{Q \notin G_i} p_Q,$$

enabling closed-form optimization of the NG parameters for arbitrary bias.

5. Applications: Entanglement and Correlation Broadcasting

Pauli cloners are integral to the study of broadcasting quantum correlations. Local and nonlocal asymmetric Pauli cloners have been analyzed for their ability to optimally redistribute entanglement and quantum discord among multiple parties. For two-qubit mixed state inputs, local application of asymmetric Pauli cloners generates output pairs whose entanglement (verified by the Peres–Horodecki criterion) depends on both input concurrence and the cloner’s asymmetry. Nonlocal cloners are strictly more effective, broadcasting to weaker initial entanglements (Jain et al., 2017).

Tasks such as $1 \rightarrow 3$ broadcasting can be approached by either successive use of $1 \rightarrow 2$ Pauli cloners or by direct $1 \rightarrow 3$ cloners; the latter are superior in broadcasting range. Broadcasting of geometric discord via Pauli cloners is fundamentally limited: for all nontrivial asymmetry, local outputs retain nonzero discord, precluding optimal broadcasting in this metric (Jain et al., 2017).

6. Key Results and Theoretical Implications

• The class of Pauli cloners, parameterized by $\ket\psi \in \mathbb{C}^{4^N}$, subsumes canonical cloners (UQCM, PCCM) and provides a continuous spectrum interpolating between universal, phase-covariant, and highly biased performance envelopes.
• Real-time adaptation of Pauli cloners using channel tomography input enables eavesdroppers to “cancel” dominant environmental errors, increasing attack fidelity, while suggesting active error introduction as a defensive countermeasure (Kerstan et al., 31 Jan 2026).
• For entanglement broadcasting, Pauli cloners define the operational boundary for clone- and broadcast-based redistribution of correlations, but for certain classes of states, economical local unitaries can extend this boundary beyond what is accessible via cloning transformations.
• The Pauli cloner framework is conjectured optimal for any $N$-qubit Pauli channel in terms of single-copy fidelity, although broad optimality for multi-copy scenarios remains open (Kerstan et al., 31 Jan 2026, Jain et al., 2017).

7. Distinction Between Cloning and Broadcasting Tasks

Cloning explicitly targets faithful (high-fidelity) replication of arbitrary unknown quantum states, as parameterized by cloning fidelities. Broadcasting, by contrast, is focused on the redistribution of entanglement or quantum correlations—fidelity to the original state is secondary. This distinction results in operational and theoretical differences: optimal broadcaster protocols may outperform cloners in redistributing correlations, and in some cases, tailored local unitaries can extend broadcasting capacity beyond what symmetric or asymmetric Pauli cloners offer. The non-uniqueness and sometimes divergent optimality criteria for these two families demarcate cloning and broadcasting as fundamentally distinct tasks in the manipulation of quantum information (Jain et al., 2017).