In briefA programmable chip combining coherent quantum light emitters with reconfigurable optics
SummaryThis paper presents a programmable quantum photonic platform that controls quantum dot single-photon emitters and reconfigurable optical elements together at cryogenic temperatures. Each active component on the chip can be addressed electrically and optically, allowing concurrent operation while preserving coherent non-classical light emission. The authors demonstrate pure single-photon generation, two-photon interference, and resonant transmission and reflection measurements of the integrated emitters.
Hanna Salamon, Mikkel T. Mikkelsen, Zhe Liu et al.
In briefA practical guide to VQE, a hybrid quantum method for chemistry and physics problems
SummaryThis article gives a structured introduction to the Variational Quantum Eigensolver (VQE), a hybrid quantum-classical method for electronic structure problems on near-term quantum hardware. It reviews key components such as problem encoding, ansatz construction, classical optimization, and measurement strategies. It also discusses practical factors affecting performance, including noise, error mitigation, and hardware-dependent circuit design, with examples from physics and quantum chemistry.
Jannis Ehrlich, Felix Rupprecht, Konstantin Lamp et al.
In briefLindblad master equation for the quantum Rabi model in deep strong coupling
SummaryThis paper derives a Lindblad-form quantum master equation for the open quantum Rabi model in the perturbative deep strong coupling regime, where the spectrum is nearly degenerate and harmonic. The derivation combines secular and singular-coupling approximations in a Schrieffer-Wolff transformed frame. The equation is benchmarked against a Bloch-Redfield equation, showing excellent agreement for the time evolution and steady states of several observables. Analytical results are also obtained in limiting cases.
In briefIs the Clifford hierarchy generalized semi-Clifford?
SummaryThe paper studies semi-Clifford and generalized semi-Clifford structures within the Clifford hierarchy, depending on hierarchy level k and qubit number n. It gives exact conditions for when the hierarchy is semi-Clifford and constructs counterexamples for larger k and n. It also introduces n_c-generalized semi-Clifford gates and conjectures their general form.
WhyGives precise characterizations and new counterexamples for semi-Clifford structures in the Clifford hierarchy, with largely complete results, though impact stays within a theoretical subfield.
In briefValue of multi-state photon-number-resolving SNSPD arrays
SummaryThe paper presents a combinatorial and Bayesian framework to quantify the information gained by improving the local photon-number resolution of multiplexed superconducting nanowire single-photon detector (SNSPD) systems. It derives a response tensor for arbitrary numbers of resolvable states, including finite efficiency and dark counts, and from it a generalized N-photon detector fidelity and Bayesian reconstruction. The authors find that higher local resolution systematically improves detector fidelity and reconstruction, relaxing the quadratic scaling required by binary multiplexed arrays.
In briefAdaptive complete quantum search for unknown solution counts: a method that combines classical sampling and quantum search to find all matching items when the number of matches is not known.
SummaryThe paper studies unstructured database search, where all elements satisfying a property must be found without knowing how many there are. It combines classical search, Grover's algorithm and generalized fixed-point quantum search into an adaptive Bayesian approach that maximizes expected success probability per unit execution time. For a known solution count much smaller than the database size, the authors prove the approach is asymptotically optimal within the considered model. Analytical results are supported by numerical simulations.
Raj Alexandru Gu\c{t}oiu, Bogdan-C\u{a}lin Ciobanu, Pantelimon George Popescu
In briefInformation leakage from Shamir secret sharing with correlated quantum resources
SummaryThe paper studies how much information leaks from Shamir secret sharing when each device uses its share to produce one qubit from an initial quantum resource that is independent of the secret. The authors prove that when reconstruction needs a large enough fraction of shares, the trace distance between leakage states for any two secrets becomes exponentially small. The results cover stationary chains and resources from finite quantum memories, and they also determine optimal exponential bases for independent device-reference pairs.
WhyProves exponentially small trace distance for leakage under correlated quantum resources and determines the optimal exponential base in the independent case.
Isaac Barouch Essayag, Aryeh Lev Zabokritskiy (Yohananov)
In briefRobust subspace designs and a unique small quantum witness: a paper on a stronger form of subspace designs and its use in quantum complexity theory
SummaryThe paper introduces robust subspace designs, a quantitative extension of subspace designs in which only a limited number of subspaces may contain directions close to a given fixed-dimension subspace. It gives a probabilistic polynomial-size construction and a non-trivial explicit superpolynomial-size construction. The concept is applied to a quantum, space-bounded version of the Valiant-Vazirani theorem, isolating a unique witness for quantum Merlin-Arthur protocols, and to related problems such as well-conditioned nullity testing.
WhyIntroduces robust subspace designs with probabilistic and explicit constructions, extending unique-witness isolation to space-bounded quantum Merlin-Arthur protocols with theoretical rigor.
Simon Apers, Roman Edenhofer, Benjamin Mathieu-Bloise et al.
In briefA metalens-based electro-optical chip for cooling levitated particles near a surface
SummaryThe paper demonstrates optical levitation of a particle in vacuum using a miniaturized hybrid electro-optical metalens. A high numerical aperture metalens combined with electrical feedback through planar electrodes controls motion and cools the particle to low phonon occupations. Reheating measurements indicate that electric field noise has no significant influence, supporting future on-chip levitation platforms near surfaces.
Pedro Rosso Gomez, Bruno Melo, Adeel Afridi et al.
In briefTwo-Sided Bounds on Ground-State Properties of Lattice Gauge Theories
SummaryThis paper proposes a framework based on a hierarchy of semidefinite programs that gives increasingly tight lower bounds on the ground-state energy of lattice gauge theories. The bounds are extended to observables such as Wilson loops and mesonic strings. The authors test the method on (1+1)-dimensional and (2+1)-dimensional Z2 gauge theories and obtain certified energy intervals. The approach is presented as a tool for studying many-body physics and certifying quantum simulator results.
WhyProposes semidefinite-program hierarchy giving certified ground-state energy lower bounds for lattice gauge theories, with numerical validation on specific models, offering solid but narrowly scoped e
Elo\"ic Vall\'ee, Ariel Kelman, Julius Mildenberger et al.
In briefAdiabatic tracking and optimal schedules for preparing Gibbs states on classical and quantum computers
SummaryThis paper presents a discrete adiabatic framework for preparing Gibbs states in classical and quantum optimisation algorithms. It proves a tracking theorem for time-inhomogeneous reversible Markov chains and derives schedules adapted to the spectral gap and energy fluctuations. Applying the discrete quantum adiabatic theorem to temperature-dependent Szegedy quantum walks, the authors state that under suitable regularity conditions the quantum cost scales as O(δ_min^{-1/2}), versus the classical O(δ_min^{-1}).]
Pedro C. S. Costa, Mauro E. S. Morales, Dong An et al.
In briefSuppressing Frequency Collisions in Quantum Processors
SummaryIn superconducting quantum processors, qubit transition frequencies that are too close together cause crosstalk, leakage, and errors dependent on neighboring qubits. The paper presents a Hamiltonian-based framework that maps two- and three-body collision mechanisms to a Max-k-Cut problem for qubit frequency allocation, and uses a cluster-stitching protocol to handle larger lattices. On the tested instances, counterdiabatic QAOA improved results near steep collision penalties, while simulated annealing achieved the lowest costs.
In briefHigh-fidelity remote graph state preparation for blind quantum computing
SummaryMeasurement-based quantum computation needs entangled graph states, but existing remote state preparation protocols only produce separable states, requiring later entangling gates on the server. This paper introduces Remote Graph State Preparation (RGSP), which prepares arbitrary graph states from a single high-dimensional photonic qudit by encoding qubits and their connectivity into the photon's phase profile. The authors report that under a reasonable noise model the fidelity does not depend on graph topology, and that a highest-weight-first mode reordering improves fidelity under cumulative fiber phase drift. They also show reduced-SWAP universal blind quantum computing, cutting the remote-state qubit count for a six-qubit Quantum Fourier Transform from 420 to 48.
In briefProbing quantum anomalous Hall transport under microwave irradiation with a topological circulator
SummaryThis paper studies how microwave photons perturb the transport of edge magnetoplasmons (EMPs) in a quantum anomalous Hall topological circulator. Pump-probe measurements show a strongly frequency-selective response, with transmission changing substantially at several pump frequencies but remaining nearly unchanged at 4 and 7 GHz. Near an exceptional point of the coupled EMP-resonator system, the pump-induced change in transmission magnitude is about twice that away from it, suggesting potential for enhanced microwave spectroscopy and sensing.
In briefUnitary RQL Equals RQL: Removing Measurements from Logspace Quantum Computation
SummaryThis paper studies whether intermediate measurements in quantum computation can be eliminated without losing power. For the standard gate set {H, T, CNOT}, the authors show that RQL and RQUL, which differ in the kind of error allowed, are equal while preserving polynomial time and logarithmic space, with zero acceptance on no-instances. The proof extends a density-matrix doubling method to general channels, including resets and classical memory.
WhyResolves an open question of Fefferman and Remscrim by eliminating intermediate measurements with one-sided error, backed by rigorous proofs with broad complexity-theoretic impact.
In briefBlack hole horizon fluctuations driven by Unruh quantum noise
SummaryThe paper studies how quantum stress fluctuations of Hawking-related fields perturb the Schwarzschild geometry of a black hole. It connects the exterior Unruh Wightman function to a finite-band stochastic reconstruction of the geometry and uses a shared Gaussian reconstruction across 2,200 realizations. The reported relative widths are 0.0185 for the local horizon radius and 0.0483 for the surface-gravity temperature parameter, with covariance derivatives explaining the unequal sensitivity.
In briefSpreading out cooled particles in hollow-core fibers
SummaryThis paper studies a silica nanoparticle held by optical interference fringes inside a hollow-core fiber, controlling its motion with adaptive techniques. After feedback cooling of its axial and radial motion, the authors widen the axial position variance in two ways. Non-adiabatic fringe suppression raises it by 11.83 (±0.7) dB relative to the cold state, and multi-pass positioning at dark fringes raises it by 13.23 (±0.5) dB, matching a Wiener stochastic model. Stronger delocalization yields non-Gaussian motional states without auxiliary traps or charged particles.
Soumya Chakraborty, Pardeep Kumar, Gordon K. L. Wong et al.
In briefMeasuring non-Abelian anyon braiding with interferometry
SummaryThis paper proposes a protocol that braids non-Abelian anyons using mobile quantum impurities with addressable internal spin states. It shows how to experimentally measure the R and F matrix elements that characterize anyonic statistics. Using Ising and Fibonacci anyons as examples, it explains how anyonic behavior appears directly in the proposed Ramsey interferometer. The authors also introduce an encoder that transfers information from conventional flying qubits into a topologically protected quantum memory.
In briefQuantum plasma weakens Casimir force between conductors
SummaryThe paper studies the Casimir interaction between two perfectly conducting plates whose gap is filled with a two-component quantum Boltzmann plasma, using a finite-temperature field theory. In the classical limit, the static longitudinal mode is Debye screened while the transverse electromagnetic sector is unaffected. Finite thermal de Broglie wavelengths modify the nonzero Matsubara response and give the leading quantum correction to the force.
Petr Brandyshev, Maksim Koniushenko, and Yury Budkov
In briefImproved Adaptive Quantum Algorithm for Partition Function Estimation with Heisenberg Scaling
SummaryThis paper presents quantum algorithms for estimating the log partition function of an n-qubit quantum Hamiltonian to additive error ε with Heisenberg scaling, while reducing calls to thermal-state purification preparation. For constant inverse temperature, the authors show a cooling schedule of length O(√n) exists and achieve Õ(n^{5/4}/ε) expected calls, improving by n^{1/4} over non-adaptive strategies. If thermofield doubles can be prepared controlled on a temperature register, both query counts drop to Õ(n/ε), which the paper states is optimal in block-encoding queries up to polylogarithmic factors.
WhyAdaptive cooling schedule with recursive-doubling identity improves query complexity by n^{1/4}, with optimality argued via rigorous complexity analysis.
Yufei Wang, Daniel Stilck Fran\c{c}a, Samuel Slezak
In briefA general method for fully characterising squeezed-light sources that accounts for impure states
SummaryThis paper presents a device-agnostic framework for characterising parametric down-conversion (PDC) squeezed-light sources, whose generated states are generally mixed and multimode due to optical imperfections. Existing methods assume pure states and can give incomplete assessments. The authors treat the source as a black box producing a mixed multimode Gaussian state and derive a lower bound on its purity. This bound can be determined experimentally with mode-unresolved measurements, without a detailed model of the nonlinear device.
Denis A. Kopylov, Michael Stefszky, Christine Silberhorn
In briefSupersinglets: a robust, symmetric class of high-dimensional multipartite entangled states
SummaryThe paper studies supersinglets, fully antisymmetric multipartite states of high-dimensional particles that remain invariant under arbitrary collective local unitaries. Using convex optimisation, the authors show their entanglement is more robust to noise than several standard entangled states under certain relaxations. They also propose witnesses based on collective-spin or randomised measurements and prove that supersinglets enable Heisenberg scaling for high-dimensional gradient sensing.
Gabriele Cobucci, Armin Tavakoli, Alexander Bernal et al.
In briefWhat It Costs to Simulate a Quantum Sentence Classifier: Energy and Compute for Near-Term QNLP
SummaryThe paper measures the classical simulation cost of a variational quantum classifier (VQC) for binary SST-2 sentiment classification across 27 configurations, comparing it with logistic regression. VQC beat its matched baseline in only 6 single-seed comparisons, and only 1 of these kept a positive advantage after reruns at two more seeds. Training was 886 to 21,127 times slower in wall-clock time than the classical fit. The study is narrow, a reproducible feasibility measurement rather than a general verdict on QNLP.
In briefTwo superconducting pairing models on the Kitaev honeycomb lattice, with microscopic kernels and parity alternation on odd-fold bond stars
SummaryThe paper constructs two superconducting pairing models on doped Kitaev honeycomb lattice, deriving their microscopic kernels rather than assuming them. Both are driven by a bond-local glue that is neutral between even- and odd-parity pairing channels, with the chirality-parity link set by bond-star geometry. A general theorem states that on odd-fold bond stars, the angular-momentum ladder of a chiral form factor alternates in parity, and the spin glue drives Chern number C=±2 while the charge kernel gives C=±4.
In briefWhen Equivalent Quantum Circuits Lose Synthesis Choices: High-Level Operations Lost During Representation Changes
SummaryThis paper examines how high-level quantum operations, such as the quantum Fourier transform and multi-controlled X gates, can be lost when circuits are serialized, exchanged as OpenQASM, or converted between compilers, leaving later compilers unable to apply a requested synthesis method. Experiments across Qiskit, TKET, Cirq, and a Munich Quantum Toolkit pipeline show that a direct OpenQASM 3 round trip removes the effect of requested synthesis in all 360 conditions, and OpenQASM 2 interchange raises the two-qubit gate count by 37.2 percent for an eight-qubit Grover circuit. The authors introduce Verified, which records high-level operations before a representation change and reconstructs them after a check, restoring requested synthesis in all 830 recovery runs.
In briefCompact analytical formula for x-ray scattering from localized excitons
SummaryThis paper presents a compact model describing inelastic x-ray scattering from spatially localized excitons. The analytical theory is combined with first-principles calculations and applied to lithium fluoride (LiF). Spectra from two approaches for the single-particle wave functions agree qualitatively with experimental data. The authors also show that a detailed description of the wave functions is essential for obtaining qualitative results.
Joris Sturm, Duc Anh Truong, Andrei Benediktovitch et al.
In briefMeasuring how entangled mixed quantum states are, and testing the measurement on a real quantum computer
SummaryThis paper studies the geometric measure of entanglement for mixed quantum states, linking it to spin correlations. The authors derive an analytical formula for rank-2 mixtures of arbitrary multiqubit pure states within a two-dimensional subspace. For equal-probability mixtures of an arbitrary multiqubit state and a Schrödinger cat state, they build quantum protocols to estimate the entanglement. The protocols were run on the AerSimulator and the ibm_marrakesh quantum computer, and the device results agree with theory within estimated error bounds.
In briefCollective quantum state preparation is nonadditive, meaning joint preparation can beat separate optimal runs
SummaryThe paper studies the resource cost of preparing many quantum states together under weighted Pauli control, where rotations are charged by angle. It derives the exact single-qubit cost and shows a four-pulse collective protocol prepares two T states more cheaply than two optimal single-copy protocols. The authors also identify Bloch-sphere regions with collective savings and give a dimension-independent Rényi-entropy speed limit.
In briefGaussian Universality and Its Breakdown in Tensor-Network Machine Learning: when tensor-network learning models become Gaussian processes and when they do not.
SummaryThe paper uses a moment-based analysis to study when Gaussian universality appears and breaks down in tensor-network learning models, focusing on matrix product states. It proves that in the large bond dimension limit, models with both local and global observables converge to Gaussian processes, with explicit finite-size bounds on higher-order moment deviations. In the large physical dimension limit, models with global observables keep persistent non-Gaussian corrections, while models with local observables still behave as Gaussian processes.
WhyProves Gaussian convergence in large bond-dimension limit with explicit finite-size moment bounds, and identifies non-Gaussian breakdown for global observables.
In briefQuantum data loading using structure learned from real signals
SummaryThe paper proposes a quantum-native data loader that learns a low-dimensional shared structure of a dataset and then prepares each signal with a fixed set of circuits controlled by a few numbers. Across seven views of five public datasets, it matches the target accuracy of the strongest structured loader at equal gate cost while using several times fewer numbers per signal. The authors report that these numbers can be inferred from a random subset, and that this subset size stayed stable in a preregistered blind replication as signal count grew sixteenfold.
Pablo Herrero G\'omez, Antonio Jimeno Morenilla, David Mu\~noz-Hern\'andez et al.
In briefThresholds of XZZX vs. rotated surface codes under correlated quantum noise
SummaryThis paper evaluates XZZX and rotated surface quantum error-correcting codes under spatio-temporal correlated noise models. The authors use the Minimum Weight Parity Factor decoder, which handles syndrome hypergraphs directly. Under strong Z-biased correlated noise, both codes show reduced thresholds, and the XZZX code displays a non-monotonic threshold dependence on bias ratio.
In briefSymmetries of 2D translation-invariant quantum codes: a framework for their structure on finite lattices
SummaryThis paper studies materialised symmetries of two-dimensional translationally invariant (TI) near-term qLDPC codes, such as bivariate bicycle codes, which can be used to build matching-based decoders. The authors decompose the symmetry spaces of 2D CSS TI codes on infinite lattices into translation-invariant subspaces and write an explicit plane-wave-like basis. They then show how to restrict this basis to rectangular periodic lattices and twisted boundaries, and give examples including the gross code.
Andrew Li, Stephen D. Bartlett, Andrew C. Doherty et al.
In briefUsing polynomial neural surrogates to speed up designing photonic quantum experiments
SummaryThis paper introduces a physics-inspired polynomial neural surrogate for PyTheus, a graph-based quantum-optics simulator, to predict the quantum states produced by experimental setups. Separate surrogates are trained for four-, six-, and eight-photon systems, achieving higher accuracy with fewer trainable parameters than standard multilayer perceptrons. The authors use these surrogates for inverse design of GHZ, W, and linear-cluster states, reporting faster design than direct PyTheus optimization for the larger systems.
In briefMetastability in Quantum Dynamics for Optimization: How Quantum Dynamics Can Stall Before Reaching a Solution
SummaryThis paper studies metastability, where a quantum optimization dynamical system stays near an intermediate state for a long time before reaching its target. Using quantum Langevin dynamics as an example, it analyzes how coherent evolution and dissipation together shape this behavior. Two-state and multi-state models, along with numerical tests on double-well and multi-well potentials, yield a spectral-separation criterion and show that changing the dissipation structure can shorten the time needed to reach the target state.
In briefFree fermion models whose operator entanglement grows as √t
SummaryThe paper studies how the operator space entanglement entropy (OSEE) of one-dimensional number-conserving free fermions evolves under random space- and time-dependent dynamics. With randomness, the OSEE of local spin operators containing a Jordan-Wigner string grows as O(√t), and the OSEE of the evolution operator also grows as O(√t). The average entropies are computed analytically by mapping the models to the symmetric simple exclusion process, with a self-averaging assumption supported numerically.
In briefJoint GKP Encoding for GHZ State Transmission over Bosonic Quantum MIMO Channels
SummaryThis paper studies sending a GHZ quantum state over noisy bosonic quantum MIMO channels using finite-energy GKP encoding. A coupled-mode model accounts for mixing, attenuation, and Gaussian noise, and the transceiver uses programmable optical meshes, MMSE gain, local GKP recovery, and joint syndrome decoding. Simulations show joint lattice shaping outperforms processing each GKP code separately. With full instantaneous channel information, GHZ-state preservation improves under coherent-basis drift, while designs using only singular values or statistics perform similarly.
Seid Koudia, Selssabil Kadid, Symeon Chatzinotas et al.
In briefFast Mixing of the SYK Model at High Temperatures (a proof of a quantum Gibbs sampler)
SummaryThe paper studies the dynamics of the Sachdev-Ye-Kitaev (SYK) model, a strongly interacting fermionic system. It shows that, with high probability over the disorder, the model admits a quantum Gibbs sampler with a spectral gap independent of system size at sufficiently high constant temperatures. This enables polynomial-time preparation of SYK Gibbs states on a quantum computer, with the result holding up to arbitrary inverse polynomial trace distance. The proof introduces a pseudo-Lindbladian approach for fermions.
WhyRigorously proves a system-size-independent spectral gap for SYK at high temperature, introducing a fermionic pseudo-Lindbladian method with notable impact on quantum simulation.
In briefRotation overrides Kibble-Zurek universality in condensate formation
SummaryThis paper studies how trap rotation affects Kibble-Zurek (KZ) universality when a two-dimensional Bose-Einstein condensate forms through a finite-time quench. Stochastic Gross-Pitaevskii simulations show that rotation-induced chiral bias drives the vortex number close to the Feynman-Onsager (FO) value, independent of quench time. Increasing rotation during the quench restores KZ scaling at low frequencies, with vortex numbers below the FO value.
Shiva Dwivedi, Kenichi Kasamatsu, Adolfo Del Campo et al.
In briefBenchmarking QEC building blocks on large quantum processors
SummaryThis paper proposes a few-sample benchmark for quantum error correction primitives, such as parity extraction and mid-circuit measurement, based on an LR-QAOA circuit with fixed linear parameters. The Hamiltonian is built from a code's check structure, and the decay of the approximation ratio with depth is mapped to an equivalent two-qubit depolarizing rate. The authors run direct and mid-circuit-measurement implementations on 10 QPUs from IBM, IQM, and Quantinuum. On IBM's ibm_phoenix, they report a correlation between the benchmark and logical-memory experiment performance.
In briefStable diamond magnetometer using dual-feedback control for long-term weak-field sensing
SummaryThis paper presents a diamond quantum magnetometer that uses a dual-feedback system to compensate for frequency shifts in double-quantum 4-Ramsey interferometry caused by thermal drift and magnetic noise. The system ran stably across a 2.8 K temperature variation. It maintained a sensitivity of 7.0 pT/Hz^1/2 over 5-100 Hz for 4 hours, reaching a noise floor of 64 fT/Hz^1/2. The authors suggest the approach could detect weak biomagnetic signals.
Yuta Araki, Takeharu Sekiguchi, Yuji Hatano et al.
In briefUsing fusion measurements to decode non-Abelian topological order
SummaryThis paper studies maximum likelihood decoding for quantum error correction in codes built from non-Abelian anyons. The authors add syndrome measurements that resolve non-Abelian anyon fusion outcomes and show that these refined syndromes improve decoder performance. They build explicit decoders for the D4 quantum double model under charge noise and obtain threshold values by mapping to classical statistical-mechanics models. They report a nonzero noise threshold below which the entire logical qudit can be recovered in a single round of measurements.
In briefLarge conductance boost in graphene nanoscrolls from interlayer-coupled states
SummaryThis theoretical study finds that rolling a graphene nanoribbon into a nanoscroll of 1.0 to 1.2 turns can raise its longitudinal conductance by nearly twentyfold. The authors attribute this to interlayer hopping between open boundaries, which turns localized edge states into one-dimensional-like conducting channels and shifts the Fermi energy. The conductance decreases as the number of turns increases, suggesting that geometry can be used to tune quantum transport.
In briefClassical observable coalescence as a diagnostic for spectral degeneracies in non-Hermitian systems
SummaryThe paper proposes a classical diagnostic that identifies exceptional points in non-Hermitian systems by observing coalescence of physical observables rather than eigenvectors. It studies a known non-Hermitian Ising chain, where mean-field equilibrium branches merge at the exceptional-point boundary with the expected square-root branch-point structure. The authors show that a degeneracy at γ=0 and η=0 is a diabolic point with no observable coalescence, so the diagnostic discriminates between degeneracy types. They support the results with exact Jordan-Wigner and Bogoliubov-de Gennes calculations and suggest a coupled gain-loss circuit as an experimental realisation.
In briefBeyond the spectral gap: rapid mixing of quantum Gibbs samplers in 1D
SummaryThis paper proves that a quantum Gibbs sampler for general one-dimensional non-commuting local spin systems mixes rapidly at every fixed positive temperature. For n sites, the KMS detailed-balanced sampler reaches trace-norm error ε in time O(log^4(n)+log(1/ε)), improving the earlier O(n+log(1/ε)) bound. The authors explain that the spectral gap can substantially underestimate equilibration speed, and they develop a framework for proving rapid mixing through spectral analysis beyond the gap.
WhyImproves mixing time of 1D quantum Gibbs samplers to polylogarithmic via a beyond-gap spectral framework with rigorous proofs, impact mainly within quantum Gibbs sampling.
In briefQuantum complexity of solving linear equations on higher-order networks, a theoretical proof that such systems are hard for quantum computers to solve
SummaryThe paper studies the quantum complexity of solving Hodge Laplacian linear systems that arise in higher-order networks, which model group interactions. The authors prove that preparing a quantum state encoding the normalized minimum-norm solution is BQP-hard, and that the associated decision problem is BQP-complete. These results offer a worst-case complexity basis for assessing quantum advantage in higher-order network analysis.
WhyProves BQP-completeness for Hodge Laplacian linear systems, giving a complexity foundation for assessing quantum advantage in higher-order network analysis.
In briefCertified quantum random numbers secured by energy limits, even against quantum attackers
SummaryThis paper studies a prepare-and-measure quantum random number generator whose security relies on physically motivated energy constraints rather than non-signalling or post-quantum computational assumptions. The authors prove a new, essentially tight lower bound on single-round von Neumann entropy against adversaries holding both classical and quantum side information. They show that using von Neumann entropy instead of min-entropy yields a major improvement, nearly matching the classical bound in many cases.
WhyProves a nearly tight single-round von Neumann entropy lower bound against quantum side-information, substantially improving prior min-entropy bounds for energy-constrained QRNGs.
Thomas A. Hahn, Efrat Gerchkovitz, Ester Yehuda et al.
In briefAn efficient Hamiltonian-based quantum algorithm for characters of the symmetric group
SummaryThe paper presents a simpler quantum algorithm that prepares a state corresponding to a column of the symmetric group's character table, using a sequence of Hamiltonian evolutions. It includes an implementation using only nearest-neighbor gates in 1D. The gate count depends on the chosen column, with a provable worst case of Õ(n^2.5), improving on the prior quantum Fourier transform approach's Õ(n^3). The authors also generalize the protocol to a quantum character transform and discuss an application to entanglement entropy in symmetric orbifold conformal field theories.
In briefReal-space interference caused by quantum metric in crystals
SummaryThis paper shows that in crystalline systems, the quantum metric distribution of an isolated band can produce real-space interference through the evanescent tails of localized bound states. Quantum metric hot spots in momentum space set complex wavevectors, and multiple hot spots create oscillating density patterns. The effect is most pronounced in (nearly) flat bands. The authors propose observing it using a Josephson scanning tunneling microscope via tip-position dependence of the Josephson critical current.
In briefA momentum-space path integral method for non-Hermitian symmetry breaking
SummaryThis paper develops a momentum-space version of the quantum-classical correspondence for non-Hermitian symmetry breaking, deriving a trace formula and quantization condition. It shows that whether each eigenvalue is real or complex depends on the symmetry of its semiclassical orbit, with the form of the quantization condition set by orbit topology. As an application, it studies boundary-induced spectral transitions of Bloch oscillations in a finite non-Hermitian lattice.
In briefUniversal Improvement of Quantum Channel Fidelity Using Entanglement Assistance
SummaryQuantum communication usually needs encoding and decoding across many uses of the same channel, but decoherence can make each use different. The paper presents an entanglement-assisted encoding and decoding strategy that yields a channel with higher worst-case fidelity than all n given channels. The improvement is universal under mild assumptions on initial entanglement fidelities and is optimal for depolarizing channels. The idea is also extended to classical channels using shared randomness and to general quantum resource theories.
WhyProposes a universal entanglement-assisted scheme for nonidentical channels, with optimality and uniqueness proofs; novel and rigorously supported theoretically.
In briefLimits on the source of a universal 1/f² photon-count spectrum at baseband
SummaryResearchers ran 11.6-day photon-counting experiments over 1.5 years using many light sources with different statistics and spectra. Every baseband photon-count spectrum followed the same 1/f² form across 1×10⁻⁶ to 5×10⁻⁴ Hz. The fluctuations lasted from hours to days and are argued not to come from source, detector, or optical-system noise. The physical origin remains unresolved, but the counts statistically resemble a doubly stochastic Poisson process with slowly varying random intensity.
In briefTransformer-Based Inference of Lindblad Dynamics from Quantum Time Series
SummaryThis paper introduces a Transformer model that takes raw time series of Pauli expectation values to infer unknown Hamiltonian parameters and dissipation rates in open quantum systems. It removes the need for hand-crafted features and extends to multiple dissipation channels and realistic measurement noise. The authors report consistently high reconstruction accuracy across all tested scenarios.
In briefClassical estimation and preparation of ground state energy via the truncated power method
SummaryThis paper studies the Guided Local Hamiltonian Problem, which estimates and prepares ground state energy given a guiding state with good overlap with the true ground state. It shows the problem is classically tractable when the true ground state has polynomially bounded ℓ1-norm. The authors prove the Truncated Power Method has runtime polynomially comparable to quantum algorithms in relevant parameters, except for an extra factor of ‖ψ‖₁². Hence, when this norm is polynomially bounded, the method dequantizes quantum algorithms for this problem.
In briefCoherence of Gaussian quantum states depends on the chosen frame
SummaryThis paper describes Hilbert space coherence using a tight frame of all canonical coherent states instead of an orthonormal basis. Gaussian states are simpler to describe in this coherent-state frame than in the Hermite-Gauss basis. The authors derive an explicit formula for the frame-dependent coherence of mixed Gaussian states in terms of the covariance matrix determinant and trace. They also construct discrete-variable states that rapidly converge to each mixed Gaussian state as the Hilbert space dimension grows to 15 or more.
SummaryThe paper shows that a many-body wavefunction exponential in R, where R squared equals the sum of squared interparticle distances, solves Schrödinger's equation for N particles with a potential proportional to 1/R. It also finds a set of translationally invariant rotational eigenstates of angular momentum.
In briefDecoherence erases energy locked by entanglement in multi-cell quantum batteries
SummaryThe study examines how charging a two-cell quantum battery can lock part of its stored energy in correlations between the cells, energy recoverable only through collective operations. This correlation ergotropy closely tracks logarithmic negativity, indicating the locked energy comes from genuine entanglement rather than coherent dynamics alone. Decoherence and dephasing erase this energy instead of releasing it, so extraction must occur while correlations persist.
In briefSelf-sustaining quantum memories built from classical random processes, offering a framework for quantum information storage without active error correction.
SummaryThe paper establishes a general equivalence between CSS-based quantum memories under Pauli noise and classical memories, using associated probabilistic processes. For asynchronous Lindbladian dynamics, it identifies a broad class where the quantum memory problem reduces to memory in continuous-time classical Markov dynamics. It also introduces syndrome-based probabilistic cellular automata and rigorously proves that the 4D toric code with Toom's rule is a quantum memory under local Pauli noise below a constant threshold.
WhyEstablishes a general classical-quantum memory equivalence and gives the first rigorous proof that 4D toric code with Toom's rule is a quantum memory under local Pauli noise.
In briefProvable quantum speedups in communication complexity that can be checked efficiently
SummaryThe paper proves optimal quantum-classical separations for search problems in communication complexity. It introduces Pelagic Fourier Fishing, where a quantum one-way protocol uses n qubits while randomized two-way protocols need Ω(2^n) bits. It also presents efficiently verifiable variants and a family of problems giving an n versus Ω_d(n^d) separation for each fixed d≥2.
WhyIntroduces a new total search problem with provably optimal quantum-classical communication separations, a rigorous proof, and significance for quantum advantage research.
In briefOptimization geometry of QAOA and VQE: local versus global search
SummaryThis paper studies the optimization landscape of variational quantum algorithms such as VQE and QAOA, focusing on local-minimum counts and quality differences among local-search endpoints. It compares multi-start BFGS with adaptive differential evolution (jSO) under matched function-evaluation budgets. Parameter tying produces more repeated local structure and larger quality gaps, where jSO outperforms multi-start BFGS. The authors conclude that the number of local minima alone does not determine whether global search is advantageous.<br>
Vojt\v{e}ch Nov\'ak, Ivan Zelinka, Silvie Ill\'esov\'a et al.
In briefNonlinear quantum spin dynamics without entanglement: a variational method extending classical atomistic spin dynamics to quantum spins
SummaryThe paper proposes a variational product-state framework that extends classical atomistic spin dynamics to quantum spins of arbitrary size s, allowing both pure and mixed local states. It excludes intersite entanglement while retaining local quantum effects from the nonclassical structure of density operators for s≥1. The authors analyze why quantum Landau-Lifshitz and Landau-Lifshitz-Gilbert dynamics are inequivalent and when they match under a common time rescaling, and they note that the method allows O(N) simulations of spin dynamics in materials.
In briefSpin-orbit-induced gauge field and spin Hall effects in light fluids
SummaryThis paper studies quasi-paraxial light fluids propagating in nonlinear Kerr media. The authors show that intrinsic spin-orbit interaction of light, encoded in Maxwell's equations, produces an effective gauge field whose sign depends on helicity and magnitude on nonlinear refractive index variations. Its effects are illustrated for flow past an obstacle and for the cumulative precession of a wave packet in a focusing fluid.
In briefA perturbative framework for non-Gaussian, nonunitary environments with nonlinear coupling
SummaryThe paper presents a perturbative framework for systems quadratically coupled to bosonic environments, which may be single- or multimode and have non-Gaussian, nonstationary initial states. Using linear evolution in a doubled operator space and a characteristic-function representation, it expresses ordered multitime environmental correlation functions via derivatives of the initial characteristic function. These correlators are used to recursively build the time-convolutionless generator at any finite order. A qubit coupled to a damped mode through a two-photon interaction serves as an illustrative example.
In briefPerformance of Multiparty GHZ-State Quantum Conference Key Agreement over Optical Fiber
SummaryThis paper studies the finite-key performance of a multiparty quantum conference key agreement system that distributes GHZ states over optical fiber. Each link is modeled as an erasure-Pauli channel extended with detector dark counts, and closed-form error-rate expressions are derived. The local bit-error rate depends only on transmissivity and dark counts, while the global phase-error rate accumulates across links and degrades faster with distance and more parties. Optimizing the parameter-estimation probability raises the secret-key rate and extends the maximum secure distance.
Dipanjan Bera, Ramniwas Meena, and Neel Kanth Kundu
In briefCorrespondence between walks and zeta functions for quantum walks on path graphs, a pre-geometric structure from a generalized reflection principle
SummaryThe paper establishes a local Konno-Sato correspondence for reflected random and Szegedy walks on a finite path graph. A generalized reflection principle using explicit involutions yields exact zeta factorizations into a boundary factor and local factors indexed by the infinite dihedral group. After symmetrization, classical and quantum local factors correspond for each group element, and the identity contribution determines the normalized thermodynamic limits. Under a double scaling of asymmetry and spectral parameter, the quantum factorization yields a massive Dirac-type limit.
In briefTelling quantum states apart using only forward oracle calls
SummaryThis paper studies quantum state distinguishing when an oracle can only be applied forward, with no inverse access, to tell apart two states separated by trace distance ε. The authors show, via a direct hybrid argument, that Ω(ε^-2) forward queries are needed in high dimension even when both states are known. They also give a dimension-independent algorithm for a continuous-time variant that uses quantum Zeno dynamics to restrict evolution to a two-dimensional subspace, with total forward evolution time O(ε^-1).
WhyProves an Omega(eps^-2) forward-query lower bound even for known states via a direct hybrid argument, and gives a dimension-independent O(eps^-1) continuous-time algorithm via Zeno dynamics; a substan
In briefMinimax quantum state tomography when the state lies in a known trace-distance ball
SummaryThis paper studies quantum state tomography when the unknown state lies in a trace-distance ball of radius R around a known reference state. For arbitrary collective measurements, it determines the minimax expected risk and confidence radius up to universal constants. It also gives the copy complexity needed to reach a given accuracy, with both bounds governed by a quantity depending on the spectrum and radius.
In briefShallow quantum circuits simulating local open-system (Lindblad) dynamics
SummaryThe paper proposes a quantum algorithm for simulating time-independent, finite-range Lindblad (open quantum system) dynamics on a lattice, allowing noncommuting jump operators. For time T≥1 and diamond-norm error at most ε, the circuit depth is O(T·polylog(nT/ε)) and the gate count is O(nT·polylog(nT/ε)). The authors also prove unconditional worst-case lower bounds, including Ω(nT) gates and Ω(T) depth, and a depth lower bound of Ω(log n/log log n) at fixed time and small fixed error.
WhyExtends HHKL spatial decomposition to open systems via shared unitary dilation, with lower bounds matching the upper bounds' time and volume dependence.
In briefLifted surgery for non-Abelian two-block group-algebra codes
SummaryThis paper extends lifted code surgery, a method that measures logical operators of quantum LDPC codes using a merged code, from Abelian to non-Abelian two-block group-algebra codes. The authors classify the block-affine automorphisms of these codes and find that most apparent non-Abelian gains vanish when measured against the full symmetry group. They prove the gain is at most the index of a largest Abelian subgroup, and report codes and circuit-level simulations with Relay-BP decoding in which the non-Abelian gadget is as reliable as the Abelian gadgets while using two to three times fewer rounds.
In briefA structured AI-agent workflow for applied quantum computing research, with a worked chemistry example
SummaryThe paper introduces the Quantum Agentic Operating System (QAOS), a structured multi-agent system for quantum and computational research projects that splits work into constrained, sandboxed tasks. It is demonstrated on a quantum chemistry prototyping task: modeling a proton-transfer problem using sample-based quantum diagonalization (SQD), with FCI and DFT as classical references. The authors compare QAOS against baseline agents, reporting that the baselines silently ignored explicit instructions and showed large run-to-run variability.}
Dikshant Dulal, Maxence Grandadam, Maciej Koch-Janusz et al.
In briefCan neural networks learn structure-preserving maps? A latent-to-Hilbert embedding case study
SummaryThe paper asks whether a neural network can learn a structure-preserving map from an autoencoder's compressed latent space to a Hilbert-space representation. Experiments use an 8-dimensional bottleneck on MNIST and n-qubit product-state targets built from PCA-based angle encoding. The learned map is well approximated by a linear function over the observed latent distribution, and kernel ridge regression with an RBF kernel outperforms a tuned MLP on inner-product error.
In briefFaster randomized compilers for simulating open quantum systems under Lindblad dynamics
SummaryThis paper develops first- and second-order randomized sampling algorithms for simulating Lindbladian dynamics, covering both time-independent and time-dependent cases. For time-independent Lindbladians, the second-order correction improves the precision dependence from ε^-1 to ε^-1/2 in the number of time slices. Because the second-order maps need not be CPTP, they are used to estimate observable expectation values rather than to prepare output states deterministically.
