We study whether local forgetting in finite-dimensional quantum memory can be composed into attenuation of information across a network. We first prove a reference-stable contraction for repeated partial-SWAP collisions between a system qubit and a memory qubit that is depolarized between collisions: for two joint system--reference states with the same reference marginal, the trace distance contracts over a two-collision block by a factor at most $1-\epsilon$, with $\epsilon=(1-\lambda)^2\sin^2(2\theta)$, uniformly in the reference system and its dimension. This estimate cannot simply be iterated across a branching structure, because the reference marginals required by the hypothesis are not preserved after the first stage. We therefore analyse one explicit branching channel ($\theta=\pi/8$, $\lambda=4/5$) through its numerically recorded Pauli transfer tensor and derive conditional exponential attenuation bounds with rate at most about $0.73$ per generation. These bounds are conditional on the correctness of the recorded transfer tensor and on a depth-three base case. Both have numerical support, but neither has yet been rigorously certified. We also prove, unconditionally, a single-path lower bound showing that the contraction coefficient of the depth-$r$ channel is at least $g^r$ with $g\approx0.68$. The channel is nearly single-path, so it only weakly tests composition, and we identify the missing ingredients for a discriminating test: symmetric children and reuse of memory. A final section reports finite-chain numerical diagnostics and a conjectured probe-mean formula for a reused-memory model; no theorem is claimed there. No Lieb--Robinson-type or spatial attenuation bound is established. Relationship to the Earlier Preprint The present work develops a substantially revised research program building on the earlier preprint, Lieb–Robinson Bounds and Entanglement Limits in Quantum Dynamics with Finite-Dimensional Memory, published on Zenodo on April 26, 2026. The earlier manuscript proposed a general memory-dependent Lieb–Robinson bound, including an effective propagation velocity and associated correlation estimates. The current work replaces that proposed theorem-level result with a mathematically established local result: reference-stable trace-distance contraction under repeated partial-SWAP collisions with depolarized finite-dimensional memory. Beyond this local result, the manuscript investigates whether such contraction can be composed into spatial attenuation. A branching-channel transfer-tensor framework is developed, with attenuation estimates stated conditionally on outstanding verification requirements. A separate shared-memory extension remains exploratory. The distinction between established, conditional, numerical, and conjectural results is made explicit throughout. In particular, the present manuscript does not claim to establish a general Lieb–Robinson-type or spatial attenuation bound. The overarching research question—whether local reference-stable forgetting can be composed into a general spatial attenuation bound for coupled system–memory networks—remains open.
We study whether local forgetting in finite-dimensional quantum memory can be composed into attenuation of information across a network. We first prove a reference-stable contraction for repeated partial-SWAP collisions between a system qubit and a memory qubit that is depolarized between collisions: for two joint system--reference states with the same reference marginal, the trace distance contracts over a two-collision block by a factor at most 1−ϵ, with ϵ=(1−λ)2sin2(2θ), uniformly in the reference system and its dimension. This estimate cannot simply be iterated across a branching structure, because the reference marginals required by the hypothesis are not preserved after the first stage. We therefore analyse one explicit branching channel (θ=π/8, λ=4/5) through its numerically recorded Pauli transfer tensor and derive conditional exponential attenuation bounds with rate at most about 0.73 per generation. These bounds are conditional on the correctness of the recorded transfer tensor and on a depth-three base case. Both have numerical support, but neither has yet been rigorously certified. We also prove, unconditionally, a single-path lower bound showing that the contraction coefficient of the depth-r channel is at least gr with g≈0.68. The channel is nearly single-path, so it only weakly tests composition, and we identify the missing ingredients for a discriminating test: symmetric children and reuse of memory. A final section reports finite-chain numerical diagnostics and a conjectured probe-mean formula for a reused-memory model; no theorem is claimed there. No Lieb--Robinson-type or spatial attenuation bound is established. Relationship to the Earlier Preprint The present work develops a substantially revised research program building on the earlier preprint, Lieb–Robinson Bounds and Entanglement Limits in Quantum Dynamics with Finite-Dimensional Memory, published on Zenodo on April 26, 2026. The earlier manuscript proposed a general memory-dependent Lieb–Robinson bound, including an effective propagation velocity and associated correlation estimates. The current work replaces that proposed theorem-level result with a mathematically established local result: reference-stable trace-distance contraction under repeated partial-SWAP collisions with depolarized finite-dimensional memory. Beyond this local result, the manuscript investigates whether such contraction can be composed into spatial attenuation. A branching-channel transfer-tensor framework is developed, with attenuation estimates stated conditionally on outstanding verification requirements. A separate shared-memory extension remains exploratory. The distinction between established, conditional, numerical, and conjectural results is made explicit throughout. In particular, the present manuscript does not claim to establish a general Lieb–Robinson-type or spatial attenuation bound. The overarching research question—whether local reference-stable forgetting can be composed into a general spatial attenuation bound for coupled system–memory networks—remains open.
How much distinguishability can a noisy quantum channel preserve when its input is correlated with an untouched quantum memory? This paper gives an exact finite semidefinite program for that question for every finite-dimensional completely positive, trace-preserving channel. The complete coefficient compares joint states with the same memory marginal. The conditional coefficient instead subtracts the distinguishability already present in the memory. Both coefficients, and conditional mutual-information contraction with a separate conditioning system, equal one minus the reciprocal of an optimization in the channel’s unnormalized Choi matrix. The equivalent inverse formula retains an essential range condition, including for singular channels. The proof solves a centered matrix-Gram problem, shows that its optimization center can be positive, and constructs a signed observable lift through an erasure channel. Two weighted quadratic inequalities give the universal entropy bound. A rare coherent memory sector proves sharpness while preserving both local marginals exactly. A memory of bounded dimension suffices, including with prescribed faithful input marginals. The same program gives exact coefficients for regular modular Petz divergences whose generators have finite endpoint value and recession slope, every fixed Petz Rényi order strictly between zero and one, and quantum Chernoff divergence. Further results include a harmonic dual, an exact tensor-product law, a strict-contraction criterion, a sharp comparison with positive quantum Doeblin minorization, and explicit classical, depolarizing and orthogonal-unitary-noise formulas. Ordinary quantum mutual-information contraction is a different optimization. For qubit depolarization the paper determines its exact value on the full completely positive parameter interval. At shrinkage one half, ordinary relative-entropy, ordinary mutual-information and complete entropy contraction are respectively 1/4, 1/3 and 3/8. In higher dimension, bounds and a strict separation are proved without asserting an exact ordinary-MI formula. A discontinuity between nearly complete amplitude damping and pure reset rules out uniformly bounded-use estimation of the exact coefficient from an unknown channel, even with adaptive experiments and quantum ancillas. Thus evaluation from exact supplied Choi data does not imply stable experimental acquisition. Rare sharpness witnesses can carry vanishing absolute information; the results do not assert a communication rate, physical simulator, general recovery map or polynomial cost in physical qubit count.
In prior work by Mozumdar [1], a Lieb–Robinson-type bound for one-dimensional lat-tice systems evolving under discrete-time collision-model dynamics coupled to finite-dimensional,exponentially mixing bond memories was established.In this paper we see what survives when each bond memory is replaced by a separable, infinite-dimensional Hilbert space : the natural setting for physically realistic memory carriers such asbosonic modes, transmission-line resonators, or itinerant photonic pulses with a long but finitecoherence time. We show that the purely combinatorial, causal part of the finite-dimensionalargument (strict nearest-neighbor propagation, the light-cone counting bound) survives verbatim,since it never uses the dimension of the memory space. The mixing part of the argument doesnot survive verbatim: two structural facts used silently in the finite-dimensional proof : normequivalence on HM , and compactness of the unit ball of B(HM ) : fail in infinite dimensions, andboth are load-bearing. We introduce a weighted, energy-cutoff mixing hypothesis adapted from theopen-quantum-systems literature on Gaussian and quasi-free bosonic dynamics, show that it restoresa version of the memory-history contraction lemma, and sketch the resulting infinite-dimensionalLieb–Robinson bound. We flag precisely which steps are rigorous under the new hypothesis andwhich remain conjectural.
We study exact reference-preserving reset as an information-transfer task in a finite tagged-source model. A declared local-qutrit realization admits a depth-one parallel source-to-bank unitary:65 pair-local exchanges reset the complete source carrier on a pure-zero bank while transferringeach source basis label to an orthogonal bank record and each source matrix unit coherently to thecorresponding bank matrix unit. We propagate this identity through an inherited work-replacementinterface and a complete-joint error contract, deriving a channel-wise calibration theorem thatcharges coefficient, timing, axis, cross-talk, leakage and bank-dynamics errors against one globalallowance. For admitted complete inputs whose nominal post-replacement source-bank comparisonis supported on the vacuum-plus-one-excitation sector with a zero bank, a per-port peak 218J givesa sufficient reset bound below 10−4 at source length 65 and preserves the existing reader allocation,while requiring 65 local ports, an aggregate peak 65·218J, and stringent joint calibration. Conversely,an ε-reset of a d-dimensional input with a pure blank environment requires environment dimensionat least ⌈d(1−ε)⌉. At ε= 10−4 the 131-dimensional tagged source still requires 131 environmentdimensions. Combining this rank bound with circuit causal cones yields memory-dependent depthlower bounds; combining environment distinguishability with a supplied positive-constant Lieb–Robinson profile yields a conditional reset-time lower bound that is nontrivial only when the profileseparation term is positive. The results quantify a memory–locality–calibration trade-off in onedeclared architecture. They do not establish optimal hardware, practical supplier generation, anautonomous repeated-use service, or a universal tight reset speed.
Quantum repeaters incorporating quantum memory play a pivotal role in mitigating loss in transmitted quantum information (photons) due to link attenuation over a long-distance quantum communication network. However, limited availability of available storage in such quantum repeaters and the impact on the time spent within the memory unit presents a trade-off between quantum information fidelity (a metric that quantifies the degree of similarity between a pair of quantum states) and qubit transmission rate. Thus, effective management of storage time for qubits becomes a key consideration in multi-hop quantum networks. To address these challenges, we propose Q-REACH, which leverages queuing theory in caching networks to tune qubit transmission rate while considering fidelity as the cost metric. Our contributions in this work include (i) utilizing a method of repetition that encodes and broadcasts multiple qubits through different quantum paths, (ii) analytically estimating the time spent by these emitted qubits as a function of the number of paths and repeaters, as well as memory units within a repeater, and (iii) formulating optimization problem that leverages this analysis to correct the transmitted logic qubit and select the optimum repetition rate at the transmitter.
We derive an extension of the Lieb--Robinson bound that accounts for memory effects in quantum lattice systems evolving under discrete-time non-Markovian dynamics generated by local interactions with finite-dimensional memory subsystems associated with each bond. The global evolution is unitary on an extended Hilbert space Hext=⨂i=1NHi⊗⨂bondsHM(i), while non-Markovian behavior, arising from repeated coupling to the same bond memory at successive time steps, leads to temporal correlations in the system. In this work, “finite memory” refers specifically to memory subsystems of fixed finite Hilbert space dimension whose reduced dynamics satisfy an exponential mixing condition in operator norm, meaning that correlation functions decay exponentially over time, with a characteristic timescale τ. More precisely, the memory channels satisfy ∥TMs(X)−Tr(X)σM∥≤ce−s/τ∥X∥, for all memory observables X, where σM is a stationary state, and c>0 depends only on the memory dynamics. For a one-dimensional lattice with nearest-neighbor interactions, we prove that the connected correlation function, defined as Ci,j(t):=supAi=Ai†,Bj=Bj†∥Ai∥≤1,∥Bj∥≤1∣Tr[(Ai⊗Bj)ρ(t)]−Tr[Aiρi(t)]Tr[Bjρj(t)]∣, satisfies an exponential bound of the form Ci,j(t)≤C0exp[−ξd(i,j)−veff(τ)t], where d(i,j)=∣i−j∣ and ξ>0 is independent of system size and time. The effective velocity veff(τ) is explicitly bounded in terms of the local interaction strength and the memory mixing parameters, ensuring the system exhibits a finite light cone, meaning that information propagation is confined to a region growing linearly with time, with a well-defined maximum speed. In the limit τ→∞, the bound no longer guarantees a finite propagation speed, reflecting the transition to a persistent memory regime where correlations do not decay exponentially. As an operational consequence, the entanglement fidelity between two-site reduced states ρi,j(t) and a maximally entangled state satisfies Fi,j(t)=⟨Φ+∣ρi,j(t)∣Φ+⟩≤21[1+Ci,j(t)]. These results extend Lieb--Robinson theory to a class of discrete-time quantum dynamics with memory subsystems associated with each bond, enabling resummation of multi-step memory effects into an effective correlation propagation kernel that governs the decay of temporal correlations. This establishes a quantitative relationship between the timescale of memory mixing and the effective speed of information propagation in the system.
In this work, we consider one of the prototypical tasks for characterizing the structure of noise in quantum devices: estimating eigenvalues of an n -qubit Pauli noise channel. Prior work [Chen , Phys. Rev. A 105, 032435 (2022)] has proved no-go theorems for this task in the practical regime in which one has a limited amount of quantum memory; i.e., any protocol with ≤ 0.99 n ancilla qubits of quantum memory must make exponentially many measurements, provided that it is . Such protocols can only interact with the channel by repeatedly preparing a state, passing it through the channel, and measuring immediately afterward. Surprisingly, in this work we show that protocols with an extremely small amount of quantum memory achieve an exponential advantage for this task. We give a protocol that can estimate any prescribed set of eigenvalues A of a Pauli channel to error ε using only O ( log log ( | A | ) / ε 2 ) ancilla qubits, O ~ ( n 2 / ε 2 ) measurements, and O ( n 2 | A | / ε 2 ) queries to the Pauli channel. In contrast, we show that any protocol with zero ancilla—even a concatenating one—must make Ω ( 2 n / ε 2 ) measurements. We also prove that the number of queries cannot be improved significantly: with k ancilla qubits, Ω ( 2 ( n − k ) / 3 ) queries to the channel are necessary to learn all eigenvalues, even for concatenating protocols.