In briefDeterministic polynomial chaos method for simulating non-Markovian quantum systems
SummaryThis paper proposes a deterministic polynomial chaos Galerkin method for simulating non-Markovian quantum state diffusion, aiming to be easy to implement, accurate, and efficient. It combines a low-rank decomposition of bath correlations with Galerkin projection onto a polynomial chaos basis, producing a formulation solvable by standard PDE solvers. According to the paper, a single deterministic solve replaces an ensemble of N_traj trajectories, cutting cost by roughly that factor and removing sampling error.
In briefSudden mass change in an interacting quantum field in a finite system
SummaryThis paper studies how the correlation functions of a quantum scalar field evolve out of equilibrium after its mass is suddenly changed (a quench) in a finite-size system. The authors numerically solve the self-consistent equation for the Keldysh correlation function within the one-loop approximation. They observe that at a time roughly equal to half the system size, the effective mass is strongly disturbed. The effect arises from interference of counter-propagating fluctuation wave fronts inside the light cone and is caused by periodic boundary conditions.
Grachik A Simonian, Alexey G. Mikhaylenko, Andrew G. Semenov
In briefWho Can Sample Forever? Shot Noise in Quantum Linear Regression
SummaryThis paper studies quantum linear regression, which includes quantum fidelity kernels and quantum extreme learning machines, where quantum features must be estimated from a finite number of measurement shots. The authors show that finite sampling makes training ill-conditioned and more sensitive to sampling noise. For quantum extreme learning machines, the systematic prediction error decreases quadratically with the number of shots N, but increasing dataset size alone cannot remove it. If training targets are also noisy, the test error can increase with N.
Gabriele Lo Monaco, Salvatore Lorenzo, Alessandro Ferraro et al.
In briefExact value and rigidity of the 2×3 magic rectangle game
SummaryThis paper studies the 2×3 magic rectangle, a nonlocal game where two players match entries under incompatible parity constraints. The authors prove its quantum value is (1+√(2/3))/2 and classify all optimal strategies in arbitrary finite local dimensions with general measurements. They show every optimal strategy contains a maximally entangled pair of four-dimensional systems.
WhyGives the exact quantum value of the 2×3 magic rectangle game and fully classifies optimal strategies with a rigorous sum-of-squares proof.
Isaac Barouch Essayag, Aryeh Lev Zabokritskiy (Yohananov)
In briefWhether certain entangled quantum states can be separated, for Bell-diagonal states
SummaryThis paper studies a family of structured quantum states called Bell-diagonal states, asking whether they are separable. It analyzes two settings: two qutrits in the Weyl-Heisenberg basis and two ququarts in a lattice basis. The authors show that in both cases no entangled absolutely PPT states exist, and they give simple spectral inequalities that are sufficient for separability. The results depend on the chosen basis, so they do not settle whether APPT and absolute separability coincide in general.
In briefUnitary-accessible coherence-erasure distance: exact qubit result and tight qutrit bound
SummaryThis paper introduces a distance measuring the minimum unitary-control cost needed to turn a quantum state into an incoherent state without changing its spectrum. Target states are restricted to incoherent states on the same unitary orbit. The authors give an exact expression for qubits and a tight universal bound with an attaining eigenbasis for nondegenerate qutrits.
In briefA clustering measure for bipartite networks whose edges carry phases
SummaryThis paper defines a clustering coefficient for bipartite networks whose edge weights are complex numbers with phases. The measure reduces to the classical bipartite coefficient when phases vanish and becomes negative when alternative routes cancel. The authors show that the phase accumulated around a four-node square is the smallest gauge-invariant carrier of phase information, and that the coefficient can be computed for all nodes from one sparse matrix product. Closed-form results for uniformly distributed phases are derived and numerically verified, and phase disorder is shown to increase coherent distance.
In briefEmergent SSH physics and localization in a cavity-atom system beyond the rotating-wave approximation
SummaryThe paper studies a cavity-atom system beyond the rotating-wave approximation by mapping its dynamics onto an effective tight-binding model in the excitation-number basis. For large photon numbers, the lattice becomes a dimerized SSH-type chain with a linear energy gradient, producing Wannier-Stark ladders and Bloch and Bloch-Zener oscillations. For small photon numbers, hopping deformation enhances collective localization and induces anomalous diffusion in selected parameter windows.
In briefNo-go theorems showing permutation symmetry alone does not yield locality, plus two exactly solvable relational history Hamiltonians
SummaryThe paper shows that permutation covariance alone does not produce a sparse metric geometry and proves several no-go results, including a lower bound on ground-state multiplicity and connectivity obstructions. It then constructs two finite relational history Hamiltonians and derives their exact gaps and ground-state counts. The results expose trade-offs between clock and locality resources, but do not establish an emergent geometric phase.
In briefRedistribution of Gaussian tripartite steering in de Sitter space
SummaryThe paper studies how Gaussian tripartite quantum steering is redistributed and directed in de Sitter space, using continuous-variable methods. It converts the Bunch-Davies vacuum into open-chart modes via a Bogoliubov transformation and derives the covariance matrix for the resulting multipartite state. The authors find that curvature generally weakens steering among accessible modes, yet two-to-one steering toward a nongravitational mode stays finite across the parameters considered.
In briefTwo ways to extend the Bloch sphere to higher-level quantum states
SummaryThe Bloch sphere geometrically represents qubit states, but extending it to pure qudit states (d>2) is not unique. This paper compares the Majorana representation, which uses multiple points, with the generalized Bloch representation, which uses a single point in a higher-dimensional space. The authors derive explicit analytical relations between the two, including a general formula for the Bloch vector in terms of a state's Majorana points.
In briefExtended Results: Absorbing Boundary Conditions as Limits of Complex Potentials
SummaryThis paper studies two methods for computing when and where a quantum particle is detected along a spatial surface or in a region: absorbing boundary conditions and complex absorbing potentials. Earlier work showed that absorbing boundary conditions arise as a limit of complex absorbing potentials, assuming a reflecting Neumann condition at the end of the potential region. Here the authors obtain absorbing boundary conditions through a different limit, using a Dirichlet condition at that end, which they argue is more practically relevant and realizable.
In briefTunable Strong Magnon Coupling and Squeezed States in a Layered Magnet
SummaryThis paper studies ultrastrong coupling between chiral and acoustic or optical magnon modes in the van der Waals antiferromagnet CrPS4. The authors split the interaction into three microscopic channels and report a normalized coupling ratio of about 0.41. They also propose extracting quadrature fluctuations and a squeezing factor of up to about 9.7 dB from thermal-noise-driven magnetization dynamics.
In briefBeyond QAOA: AI and Quantum Computing Review for Adaptive Optimization
SummaryThis review surveys how machine learning and quantum computing are combined for combinatorial optimization, organizing the literature by the decisions being learned rather than by application. From a 119-paper corpus, 67 papers were coded in detail. The authors find stronger evidence for AI-assisted quantum optimization than the reverse, but experimental controls are thin, and they introduce an M0-M5 evidence hierarchy with no convincing result at the top level.
In briefHow the choice of circuit parameters affects noise resilience in QAOA
SummaryThis paper studies how noise affects the Quantum Approximate Optimization Algorithm (QAOA), a variational method for combinatorial optimization on near-term quantum hardware. Under a Pauli noise model, the authors derive an upper bound on the total variation distance between noisy and noiseless circuit outputs, as a function of the variational parameters. Noisy simulations support this bound. The authors also propose optimization methods using error penalties and truncated search, and extend the analysis to the quantum adiabatic algorithm via the continuous limit.
In briefNonperturbative Vibrational Excitation by Arbitrary Electric Fields: A Classical Phase-Space View
SummaryThis paper presents a nonperturbative framework for quantum vibrational excitation in molecules driven by electric fields of arbitrary time dependence. For harmonic potentials with linear dipole coupling, the dynamics are exactly described by a displaced coherent state governed by a complex phase-space coordinate, with closed-form expressions distinguishing resonant from non-resonant driving. Vibrational populations follow a Poisson distribution set by a dimensionless, time-dependent phase-space displacement. The method reduces to standard time-dependent perturbation theory in the weak-field limit and remains valid in the strong-field regime where perturbative approaches fail.
Sameernandan Upadhyayula, Ji\v{r}\'i J. L. Van\'i\v{c}ek
In briefError-corrected memory and logic on a 156-qubit superconducting processor
SummaryThis paper runs a large dynamic compass code on a 156-qubit superconducting processor with a heavy-hexagonal layout. By using a noise-informed decoder and excluding the worst-performing qubits and couplers, the authors report up to 35% lower logical error probabilities for memory and up to 56% gains in measurement-based logic. They also demonstrate fault-tolerant logic via lattice surgery, with Bell fidelities raised above 97% using post-selection.
WhyCombines noise-informed decoding and excluding weak qubits on a 156-qubit device, showing memory and lattice-surgery logic gains; a substantial engineering integration on real hardware.
Campbell K. McLauchlan, Georgia M. Nixon, Julien M. Drouet et al.
In briefFrom Solid-State Spin Defects to Fluorescent Proteins: Quantum Optical Sensing for Biomedicine
SummaryThis perspective review traces how quantum optical sensing has moved from solid-state spin defect platforms toward bio-integrated systems. It covers nitrogen vacancy centers in diamond and emerging fluorescent protein based spin qubits. The authors describe advances such as photonic cavities, chip-based architectures for in vitro and in vivo studies, and nanodiamond probes for intracellular nanothermometry. The review concludes by summarizing progress and the remaining challenges for life-science applications.
In briefDeriving the dielectric constant of holographic insulators from quantum geometry
SummaryThis paper studies the dielectric properties of holographic insulators using the quantum geometry of their ground state. The authors realize the insulating state through a pseudoscalar interaction of holographic fermions and compute the fermionic Green function to extract an effective Hamiltonian. They then derive the quantum metric and Berry curvature to calculate transport properties, and obtain the dielectric constant from the imaginary part of the conductivity.
In briefSpin-outcome correlations and memory-assisted uncertainty in Bhabha scattering with an entangled spectator
SummaryThis paper studies spin-outcome correlations in tree-level Bhabha scattering, where an incident electron scatters with a positron that is entangled with a spectator electron. It compares the total memory-assisted uncertainty for three Pauli measurements with the entanglement entropy across the A|BC partition. Stronger A|BC entanglement is generally accompanied by lower uncertainty, most clearly near high-energy backscattering. The authors conclude that initial B-C entanglement can reduce this uncertainty by strengthening correlations with the spectator.
In briefNonreciprocal remote magnon entanglement via the Barnett effect with coherent feedback
SummaryThis paper proposes a scheme to generate and control nonreciprocal entanglement between remote magnons in a cavity magnonic system with coherent feedback. Two microwave cavities are driven by independent single-mode squeezed vacuum fields, with a rotating YIG sphere in one cavity and a stationary one in the other. Reversing the bias magnetic field on the rotating sphere flips the sign of the Barnett shift, producing entanglement that depends on field direction. Tuning cavity detuning and feedback phase can make entanglement appear for one field direction while vanishing for the opposite.
In briefFast fidelity decay extraction in random circuit benchmarking via classical post-processing
SummaryThe paper notes that randomized benchmarking with structured circuits gives biased short-depth fidelity estimates because incomplete ensemble mixing introduces transient modes. The authors propose a trace filter function and a one-parameter family of estimators that interpolate between standard LinXEB-type post-processing and this new method. Experiments on a Rigetti superconducting processor with one-dimensional circuits of up to 12 qubits show consistent estimates below 10 layers, where the standard estimator needs depth beyond a 40-layer window.
Ake K\"ohne, Stavros Efthymiou, Joanna Majsak et al.
In briefA quantum circuit using quantum neural networks to run flow models more efficiently
SummaryThis paper studies how to simulate wavefunction flow on quantum computers, enabling flow models to run on quantum hardware. The authors first use a QROM-based phase kickback framework to generate probability densities close to those of a conventional flow model. To reduce circuit resource costs, they replace QROM for data encoding with a trained quantum neural network. Numerical experiments show the method keeps accuracy comparable to the QROM-based approach while significantly reducing circuit resources.
In briefPower-Liouville Hermite Orbit: Shared Spectral Quantization of a Sextic Oscillator and Fractional-Power Potentials
SummaryThis paper studies a singular sextic oscillator from the finite-Hermite biconfluent-Heun hierarchy, which generates an exact four-member power-Liouville Hermite orbit. Power coordinate transformations with a Liouville amplitude map the fourth hierarchy member to four half-line Schrödinger spectral charts. The orbit includes two previously studied potentials, x^{2/3} and an inverse-square-root conditionally integrable potential. The authors derive an exact discrete sequence of quartic Sturmian eigencouplings for the sextic oscillator and confirm representative eigencouplings and node counts with independent two-sided shooting.
In briefTopological superradiant criticality in cavity-coupled spin chains
SummaryThis paper studies superradiant phase transitions in light-matter coupled systems and identifies a new type of criticality that differs topologically from known superradiant transitions. Using cavity-mediated cluster-Ising chains as an example, numerical results show that the critical line between the superradiant phase and the cavity topological phase has symmetry-protected edge degeneracies. The authors report that this criticality also appears in cavity-mediated Z_n cluster-qudit chains.
In briefQuantitative rules for parallel calibration and drift-resilient maintenance of large superconducting quantum chips
SummaryThe paper studies two factors that limit automated calibration of superconducting quantum processors as they scale: crosstalk from parallel operations and drift of control parameters over time. On a 66-qubit processor, single-qubit calibration tolerates full-chip concurrency with little fidelity loss, while CZ calibration needs a coupler-graph distance of at least four to avoid higher errors. The authors also give refresh-priority rules based on error-budget consumption, and report reducing mean single-qubit error from 0.023 to 0.011 on a 337-qubit processor.
In briefAn iterative Bayesian variational quantum eigensolver using the von Mises-Fisher distribution
SummaryThis paper presents an iterative Bayesian variational quantum eigensolver that estimates the unknown ground state, represented as a real unit vector, using a von Mises-Fisher distribution updated from Hadamard-test measurements on one ancilla qubit. It extends an earlier single-success-outcome method to condition on m success outcomes per iteration, using a closed-form posterior in the large-concentration regime. The authors prove that the overlap and concentration both remain non-decreasing for m within an explicitly characterised range, and illustrate convergence on random three- and four-qubit Hamiltonians, HeH+, and H2.
In briefDeterministic large-scale photonic GHZ states via spin echo in a quantum dot source
SummaryThe paper reports a semiconductor quantum dot device that deterministically emits indistinguishable single photons at GHz rates, forming a robust Greenberger-Horne-Zeilinger (GHZ) state of more than a dozen photons. It uses a natural spin echo effect, where electron and hole spins precess in opposite directions, and tunes the excitation rate to cancel the accumulated phase. The authors describe this as a record-length GHZ state source that could be further improved.
WhyUses natural spin echo in a quantum dot to deterministically produce a record GHZ state of over a dozen photons; novel, though experimental detail in the abstract is limited.
In briefRobust quantum key distribution despite flaws in state preparation and measurement
SummaryThis paper analyzes the security of quantum key distribution when encoding, intensity modulation, and their characterization are imperfect. It uses the operator dominance method to estimate the secret key rate tightly. Simulations using parameters from commercial QKD systems show key rates close to ideal conditions, reaching 250 km in the finite-size case, and the method is also shown to work for measurement-device-independent QKD.
In briefExperimental measurement-device-independent quantum key distribution with imperfect state preparation over 300 km
SummaryThis paper reports an experiment on measurement-device-independent quantum key distribution (MDI-QKD) that accounts for state-preparation uncertainty (SPU) in the light sources. The authors use the operator dominance method to give a tight estimate of the secret key rate. They achieve secure key distribution over 303.37 km, which they state is the longest distance in experiments involving SPU.
In briefQuantum annealing of random transverse-field Ising models in 1D, 2D and 3D
SummaryThe paper studies how slowly quantum annealing must run to avoid defect formation in the random transverse-field Ising model, using the adiabatic theorem. A strong-disorder renormalization group approach treats annealing as the aggregation of strongly coupled clusters, exposing the elementary error-forming events. Numerical results in one, two and three dimensions show the error density decreases logarithmically slowly with annealing time, with dimension-dependent critical exponents, and the authors map potential defect events in the excitation energy versus control parameter plane.
Istv\'an A. Kov\'acs, Neil Shah, R\'obert Juh\'asz
In briefA short paper on thermal perturbations causing decoherence in quantum cosmology: an open quantum systems approach
SummaryThis paper uses a toy model to study how a thermal bath of cosmological perturbations can decohere the minisuperspace degrees of freedom of the universe, treating the perturbations as an environment within open quantum systems. The authors include the perturbations' backreaction on the background through an effective second-order coupling and derive a non-Markovian master equation. For a representative set of parameter estimates, they find the decoherence timescale from the thermal contribution can be shorter than the Hubble timescale, and they explore whether the resulting pointer-state structure could serve as a criterion for choosing semiclassical states.
In briefMinimal scenarios showing genuine multipartite nonlocality in a possibilistic sense
SummaryThe paper studies a strong form of genuine multipartite nonlocality (GMNL), where a quantum strategy in an n-party scenario produces outcomes from a restricted set with certainty, while any strategy using nonclassical resources shared among at most k<n parties fails with positive probability. For the simplest five-party Bell scenario, the authors use the inflation technique to show that a game with a perfect quantum strategy robustly separates from models built from bipartite resources under local operations and shared randomness (LOSR), establishing LOSR-GMNL. They also construct a family of robust LOSR-GMNL games with ε-optimal quantum strategies that become experimentally indistinguishable from probability-one-winning strategies in the limit.
In briefHeisenberg echo symmetrization: a method to reduce gate-insertion errors in quantum computing
SummaryGate-insertion error mitigation repeats gates to amplify noise so errors can be extrapolated, but this fails for non-commuting gate noise. The paper proposes Heisenberg-echo symmetrization (HES), which adds an echo inspired by the Heisenberg representation to cancel the leading error. In simulations, HES reduces absolute fidelity error by one to two orders of magnitude compared with standard gate insertion.
In briefWhen Does Interference Help Learning? Kernel Geometry as a Pre-Experimental Test for Photonic Reservoir Computing
SummaryThis paper examines how multi-photon interference affects photonic quantum machine learning. Using the kernel geometry of boson-sampling feature maps, the authors compare interference with distinguishable-particle statistics in a task-independent way, finding that interference spreads feature variance across more effective dimensions and creates a large geometric separation between the kernels. The separation is learnable on adversarially constructed tasks under experimental sampling budgets, but about an order of magnitude smaller on natural tasks, depending on kernel-task alignment.
In briefMagnon-cavity-enhanced quantum magnetometer using nitrogen-vacancy spin ensembles
SummaryThis paper proposes a chip-scale magnetometer using nitrogen-vacancy centre ensembles, built on a one-dimensional YIG/CoFeB magnonic crystal with a structural defect acting as a passive cavity and integrated permanent magnets for bias. Simulations show a defect-localised spin-wave mode 40 MHz from the NV zero-field splitting. Modelling predicts a projected sensitivity of 93 pT/√Hz from combining three readout edges, with an optimum near 0.2-0.4 MHz NV-magnon coupling.
In briefPhotonic chips for universal quantum computing: a review of progress and outlook
SummaryThis review surveys recent advances in integrated photonic quantum chips for universal quantum computing, covering both discrete-variable and continuous-variable approaches. It discusses fault-tolerant architectures and leading integrated platforms. The authors identify the key challenges to scaling photonic systems toward full fault tolerance. They also outline future research directions for moving from proof-of-principle demonstrations to scalable quantum microprocessors.
In briefSensitivity of quantum annealing gaps and bottleneck shifts in sparse constraint-preserving drivers
SummaryThis paper studies how different coupling graphs that keep the same feasible sector can produce different minimum energy gaps during quantum annealing. The authors derive a sensitivity measure of the normalised minimum gap to exchange terms in fixed-cardinality Laplacian drivers and use it to find locally beneficial driver changes. Experiments show the sensitivity reliably predicts infinitesimal improvements but becomes less reliable for finite edge replacements, because improving the current bottleneck can expose a competing minimum. A tied-minimum max-min update is proposed as more reliable in the tested oracle setting.
In briefQ-Capsule: A localized capsule-based quantum neural network that reduces barren plateau training problems
SummaryVariational quantum algorithms often struggle to train because gradients vanish in large circuits, a problem known as barren plateaus. This paper proposes Q-Capsule, which combines register partitioning, local readout, sparse inter-capsule coupling, trainable data re-uploading, and QFIM-guided adaptive depth growth. The authors report that gradient variance stays stable as register width grows, while globally entangling baselines decay exponentially. Q-Capsule reaches 98.1% accuracy on binary classification and 97.7% on four-class classification, using about 73% fewer two-qubit gates than the fixed-depth model on the multiclass task.
Awal Ahmed Fime, Tasfia Zaman Samiha, Saika Zaman et al.
In briefMarkov length can diverge even when the universal physics is spatially Markovian
SummaryThe paper studies conditional mutual information (CMI), a measure of spatial non-Markovianity used to define mixed-state phases. It shows that states flowing to the same renormalization-group fixed point can still have qualitatively different CMI. Starting from a critical Gibbs state of a finite-range classical Hamiltonian, a detailed-balance-breaking perturbation that is RG-irrelevant generates power-law CMI tails, but these are cutoff-suppressed and vanish in the continuum limit.
WhyShows states at the same RG fixed point can have distinct CMI decay, with solvable models and a 1D criterion; novel theory, narrower practical reach.
In briefExact polariton-condensate states from nonlinear Stark coupling in cavity arrays
SummaryThe paper studies coupled optical cavities, with or without two-level atoms inside. Using a building-block construction enabled by nonlinear Stark coupling, it identifies exact many-body eigenstates that stay analytically tractable across lattice geometries. Depending on their energy, these states act as quantum scars in the chaotic spectrum or as ground states showing off-diagonal long-range order and forming polariton condensates.
In briefHigh-rate concatenated quantum error-correcting codes for qudits, which are higher-dimensional quantum units.
SummaryThis paper extends many-hypercube (MHC) codes, a type of high-rate concatenated quantum code, from qubits to qudits of prime dimension q. The authors construct level-2 and level-3 codes and generalize a level-by-level minimum-distance decoder. Under an X-error model, the error threshold rises with q, reaching 10.0% at q=13, nearly double the 5.1% for qubits. The analysis links this advantage to the entropy of the noise channel.
In briefRetromorphic testing method for verifying quantum compiler optimization passes
SummaryThe paper analyzes unit tests for quantum compiler passes in four frameworks (PennyLane, Qiskit, Cirq, pytket) and finds they mostly rely on program-content and metric assertions with small, shallow test circuits. The authors propose an automated testing methodology based on retromorphic testing and Hadamard test principles, implemented as the RetroQ framework. Applied to PennyLane and Qiskit, it reproduced existing bugs and uncovered new defects, including symbolic parameter handling and commutation logic errors.
In briefLow-frequency SWAP spectroscopy reveals coherent two-level defects in superconducting qubits
SummaryTwo-level system defects cause loss in superconducting quantum circuits, and detecting them with SWAP spectroscopy previously worked only for strongly coupled defects. The authors use flux pulses only during state preparation and readout, switching them off during relaxation delays for finer resolution and longer interaction times. They observe Rabi chevrons at swap rates down to 0.1 MHz, and report a median energy-relaxation time of 10 microseconds for these coherent defects.
Joaquin Minguzzi, John Clai Owens, Oskar Painter et al.
In briefQuantum Magic and the Strong Coupling Constant: A paper linking minimal quantum magic to the strong force's strength
SummaryThe paper studies quantum magic production in six flavor-diagonal color-singlet quark scattering channels at tree level, including photon, gluon, W/Z and Higgs exchange. Varying only the strong coupling α_s, the authors report that the finite-α_s magic minimum in the top channel matches the MS-bar value of α_s to within 5% from 1 TeV to 10^9 GeV. Higgs exchange involving the large top Yukawa coupling is essential to this agreement. The authors conjecture that fundamental interaction parameters reflect a principle of quantum computational efficiency.
In briefLower bound on overhead for addressable surgery on sparse stabiliser codes
SummaryThe paper shows that for any stabiliser code there is a logical subspace whose measurement requires Ω̃(k²) ancillary qubits when using a sparse surgery ancilla. For CSS codes, this hard subspace can be chosen as generated by all-X or all-Z operators. Consequently, for codes with large k, constant-overhead ancillas can realise only a small fraction of possible measurements.
WhyProves a $\tilde{\Omega}(k^2)$ ancilla lower bound for some logical subspace under sparse surgery, a meaningful theoretical limit.
In briefAttainable Boundaries vs. Operational Equivalence: Dual Benchmarks for Quantum Resource Assessment
SummaryThe paper proposes a framework for judging how useful arbitrary quantum states are, relative to the states that current experimental setups can actually produce. It uses two reference sets: states reachable by free operations, and a set defined by numerical optimization for operational equivalence. The authors give a strategy-dependent benchmark and a computable strategy-independent lower bound on relative performance. For entanglement, they show that distillable entanglement can be fully recovered through extraction and concentration when the reference set is restricted to pure states.
In briefQuantum simulation of hyperbolic geometry using superconducting circuits
SummaryThe paper proposes implementing hyperbolic geometry as a quantum simulation within the parameter space of a superconducting quantum circuit. The authors build a Hyperbolically Driven Quantum System by mapping hyperbolic properties onto dynamically controllable Hamiltonian parameters. They probe the system's response under geometric quantum driving to show how the parameter space's intrinsic geometry shapes observable quantities, and extend the framework to ensemble simulation.
In briefFast 2 Surgery: fast surgery on QLDPC codes with Õ(n(k+d)) ancilla overhead
SummaryThe paper shows that the fast surgery framework of Baspin-Berent-Cohen can be implemented with at most Õ(n(k+d)) ancilla qubits, improving on the previously achievable O(n·k·d) overhead. This makes fast surgery cost comparable to the Õ(nk) overhead of regular surgery. The authors build the surgery ancilla by starting from a dense but expanding system and then applying expansion-preserving sparsification methods from gauging and code embeddings.
In briefSimulating φ⁴ field theory with a superconducting transmission line
SummaryThis paper proposes a superconducting transmission line for analog quantum simulation of (1+1)-dimensional φ⁴ field theory. The line uses fluxonium-like circuit elements whose continuous, unbounded phase variables encode a scalar field. Tuning circuit parameters gives a single-well regime and a double-well regime, the latter with degenerate vacua, enabling studies of particle scattering and kink-antikink collisions.
Ilan T. Rosen, Neill C. Warrington, Max Hays et al.
In briefExperimental demonstration of a photonic quantum algorithm for computing excited state energies of molecules and spin chains.
SummaryThe paper realizes the subspace-search variational quantum eigensolver (SSVQE) on a photonic platform by encoding orthogonal qudit states in the orbital angular momentum of single photons and optimizing a shared variational circuit. The authors estimate excited state energies for a four-dimensional HeH+ molecule and an eight-dimensional Heisenberg chain. They conclude that photonic qudits offer a resource-efficient route to excited state quantum simulation.
In briefFaster Entanglement in Atoms via Cavity-Mediated Nematic Dynamics
SummaryThis paper studies generating many-body entanglement in atomic ensembles coupled to a cavity, which could enhance sensing devices. Prior approaches for atoms within a single hyperfine manifold require strongly off-resonant laser fields and have very slow dynamics. The author shows equivalent squeezing dynamics can be achieved much closer to resonance, generating entanglement over a hundred times faster than typical experiments. The Hamiltonian can also speed up creating maximally entangled states via adiabatic passage.
In briefStructured leakage in OAM-encoded qubits, studied with distributed quantum feature extraction
SummaryThis paper models orbital angular momentum (OAM) light modes as logical qubits that travel through noisy channels, where turbulence appears as leakage out of the logical subspace. The leakage depends systematically on turbulence strength and OAM order. Building on this, the authors propose a distributed quantum feature-extraction method that pulls out turbulence-robust features, and report simulations and experiments supporting it across a range of conditions.
In briefA flash-based relativistic collapse model for quantum fields
SummaryThis paper proposes a relativistic version of spontaneous collapse models, which aim to solve the quantum measurement problem. It addresses issues such as superluminal signaling, dependence on spacelike foliations, energy production, and vacuum stability, at the cost of introducing a new space of physical states. The framework is applied to a free massive scalar field, giving finite particle-number and energy-increase rates at leading order in the long-time regime.
In briefGeometry-optimized hyperbolic codes for modular fault-tolerant quantum architectures
SummaryThe paper presents finite families of hyperbolic surface codes that exploit the efficiency scaling achievable by surface codes with bounded local geometry. By optimizing periodic identifications, the authors double the code distance at fixed qubit count, encoding rate, and parity-check weight, increasing efficiency fourfold. A topology-aware algorithm partitions these codes into bounded planar modules to reduce long-range communication demands. Circuit-level simulations show that tripling long-range gate error probability mildly lowers but does not eliminate the estimated error-correction thresholds.
In briefSeparating ultrastrong Rabi splitting from quantum vacuum dressing in a chiral cavity
SummaryThis paper studies cyclotron resonance coupled to a single circularly polarized cavity mode, tuning the magnetic field to control whether the cavity field co-rotates or counter-rotates with the matter excitation. Co-rotation gives resonant energy exchange, a large Rabi splitting and conventional polaritons, with the ground state remaining the bare vacuum. Counter-rotation produces anomalous pair creation and annihilation, removing the usual anticrossing while the ground state acquires squeezing and virtual excitations. The authors conclude that spectral hybridization and vacuum dressing are distinct manifestations of ultrastrong coupling.
Ayoub El-Amrani, Nasir Boubekraoui, Zakaria Mzaouali et al.
In briefQuantum geometry at arbitrary genus (I): a quantum relation linking perturbative and instanton saddles for anharmonic potentials
SummaryThe paper studies quantum mechanics with higher-genus potentials, asking whether the perturbative and non-perturbative instanton saddles stay linked by a P/NP relation as in genus one. Using complex algebraic geometry and exact WKB, the authors show that a modified Riemann bilinear identity resolves the obstruction for genus two and above. They derive explicit all-orders relations for genus-2 (quintic and sextic) and genus-3 (septic and octic) anharmonic oscillators.
In briefCan the Quantum Fisher Effective Dimension Judge a Hybrid Quantum Neural Network?
SummaryThis paper examines what the effective dimension of a quantum Fisher information matrix actually measures in hybrid quantum neural networks. The authors test it across two image datasets, four frozen backbones including a random-projection control, four circuit depths on eight qubits, and three seeds. Its correlation with test accuracy varies widely in sign and size, and it is nearly identical across backbones, so the authors conclude it reflects circuit geometry rather than model quality.
In briefUncertainty Quantification for Quantum Fidelity Estimation: A Conformal Prediction Approach to Reliable Confidence Intervals
SummaryThis paper addresses the lack of rigorous uncertainty quantification in quantum fidelity estimation by proposing a hybrid Bayesian-informed conformal prediction framework. It yields distribution-free prediction intervals and works with both classical point estimators and neural network estimators. The method is tested under finite-sample noise, tomographic incompleteness, physical model error, and out-of-distribution data. The authors report that conformalized estimators stay statistically valid where traditional methods fail, with a class-conditional variant performing best..
In briefOptimal ring focusing of vortex beams through a circular aperture
SummaryThe paper formulates diffractive ring focusing of orbital-angular-momentum beams through a finite circular aperture as a variational optimization problem. It derives an exact analytical optimal pupil and a rigorous upper bound on achievable ring intensity. The authors also compare diffractive and lens-assisted focusing and report optical experiments testing the predictions.
Roohollah Ghobadi, Carrie Rodriguez, Lukas Scarfe et al.
In briefHow noise erodes entanglement between higher-dimensional quantum particles, via negativity rank
SummaryThe paper studies how quantum noise weakens entanglement in multi-qudit systems, using the negativity witness as the entanglement measure. By analyzing eigenvalue spectra of partially transposed density matrices, it finds that higher negativity rank gives more robustness to targeted noise but less to depolarizing noise. The authors also derive noise boundaries for the APPT subspace and the maximum qutrit-qutrit APPT purity.
In briefUsing coherence to prepare quantum states and control heat flow through repeated interactions
SummaryThis paper studies how a qubit relaxes and settles into a steady state when it repeatedly interacts with a coherent environment via an energy-conserving flip-flop coupling, within a Markovian repeated-interaction framework. The authors derive an exact analytical solution that does not rely on perturbation in the environmental coherence. They find that environmental coherence can serve as a resource for dissipative state preparation, but the system's purity can never exceed that of the ancilla. Coherence can also modify or reverse the natural direction of heat transfer.
In briefScaling quantum optimization to thousands of qubits via distributed quantum sampling
SummaryThe paper addresses the Minimum Birkhoff Decomposition, which seeks a sparse weighted sum of matchings and has uses in network scheduling and energy trading. The authors combine single-layer QAOA sampling, Extended Fully-Corrective Frank-Wolfe optimization, spectral graph partitioning, and greedy feasibility repair. They run the pipeline on the 1,354-node PEGASE grid test case across IBM superconducting quantum processors. Repaired QAOA samples achieve lower residual error than simulated annealing and uniform random sampling, and spatial circuit packing reduces hardware execution time by about a factor of three.
Amana Liaqat, Ahmed Darwish, Stephen DiAdamo et al.
In briefReconstructing black hole islands from Hawking radiation using the algebraic Petz map
SummaryThis paper addresses the island proposal for evaporating black holes, which requires operators behind the horizon to be recoverable from Hawking radiation. It introduces a protocol using the algebraic Petz map and modular theory to reflect island operators into the accessible complement, transport them to the boundary, and map them into the radiation algebra. The authors also show that this algebraic derivation naturally yields Papadodimas-Raju mirror operators.
In briefDesigning Quantum Interactions: From Effective Models to Validated Predictions
SummaryThis paper presents a unified framework for building and validating effective descriptions of quantum systems, using tools such as projection methods, Schrieffer–Wolff transformations, Magnus expansions, adiabatic elimination, and Floquet theory. The quantum Rabi model and its dispersive limits serve as recurring benchmarks for identifying control parameters, virtual processes, and failure modes. The authors organize a physical prediction into three steps: encoding, effective evolution, and observable reconstruction. Companion notebooks with QuTiP implementations and convergence tests are provided.
Sinara S. Dourado, Ciro Micheletti Diniz, Gabriel P. L. M. Fernandes et al.
In briefQuantum method for predicting future event probabilities with less memory
SummaryThis paper addresses how to estimate the probability of future events, such as earthquakes or financial shocks, when those probabilities depend on growing amounts of past data. Such estimates are limited by both computation and memory, and existing quantum methods mainly reduce computational error. The authors propose a quantum algorithm that reduces systematic bias while keeping a quadratic speed-up, aimed at near-term quantum devices where quantum memory is limited.
In briefSingle-shot magic state factories for quantum LDPC codes
SummaryThis paper proposes a space-time-efficient factory that produces magic states, the resources needed for non-Clifford gates, within high-rate quantum LDPC codes. It combines transversal non-Clifford gates from tricycle codes with single-shot four-dimensional balanced product quadcycle codes, using an escape mechanism that teleports postselected logical magic states into larger-distance codes. Under a standard circuit-level noise model at physical error rate 10^-3, the authors report CCZ states at logical error rates down to 10^-9, with 10 to 40 times lower space-time cost than prior methods.
WhyCombines tricycle transversal non-Clifford gates with single-shot quadcycle codes via an escape mechanism for magic state factories. Novel design; evidence is circuit-level noise simulations with mode
In briefNonlinear Feedback for Optimizing Josephson Circuits via a Kerr-Reversal Amplifier
SummaryThis paper addresses the high computational cost of optimizing Josephson-based nonlinear microwave devices. It introduces a nonlinear feedback extension that passes information from the pumped response back into the initial linear optimization stage. The approach is studied on a Josephson traveling-wave parametric amplifier with a reversed-Kerr architecture, whose model is validated against experimental gain measurements. The difference between linear and pumped responses of a reflection-based metric is proposed as a candidate feedback observable.
Emanuele Palumbo, Alessandro Alocco, Andrea Celotto et al.
In briefQuantum jumps of sound: detecting discrete phonon energy transitions in a vibrating resonator with a superconducting qubit
SummaryThe researchers coupled a superconducting qubit dispersively to a nanomechanical resonator and performed repeated quantum nondemolition measurements of its phonon number. The resonator reached a mechanical lifetime of 2.1 milliseconds and a dispersive shift of 328 kilohertz per phonon. Single-phonon states were heralded with 85% fidelity, and quantum jumps between the first excited and ground states were observed.
WhyFirst repeated quantum nondemolition phonon-number measurements with a qubit observing single-phonon quantum jumps; strong experimental evidence, impact mainly within quantum acoustics.
Takuma Makihara, Erik Szakiel, Matthew P. Maksymowych et al.
In briefQuantum simulation of material-specific Hubbard models using alkaline-earth atoms
SummaryThis paper proposes an integrated quantum-simulation architecture using fermionic alkaline-earth atoms that gives programmable control over the key elements of multi-orbital material Hamiltonians. It uses the four-band low-energy Hamiltonian of the bilayer nickelate La3Ni2O7 as a benchmark. Self-consistent calculations indicate that parameters from first-principles modeling can be reproduced in a single experimentally accessible lattice configuration.
Henning Schl\"omer, Reuben R. W. Wang, Bo Xing et al.
In briefQuantum Networking at Quantum Computing Speed: Kilohertz Entanglement Between Different Quantum Systems
SummaryThe paper entangles a trapped barium ion with a silicon-vacancy center in a diamond nanophotonic cavity. A photon from the ion is converted in wavelength and encoding, reflected off the cavity system, and detected, and one detected photon signals successful entanglement. Including all conversion losses, the average generation rate is 1.03 kHz, about one Bell pair per millisecond, matching the cycle time of planned trapped-ion processors.
WhyEntangles a trapped ion with a diamond SiV center at about 1 kHz, four times faster than prior ion-ion photonic links, shown experimentally.
Lukas Hartung, Pierre Barral, Noah Glachman et al.
In briefEntanglement entropy of scattering between one particle type and the rest
SummaryThe paper constructs the entanglement entropy for a general scattering process between one particle species (A) and the remaining final-state particles. Starting from a pure initial momentum eigenstate, it shows that the reduced density matrix of particle A is block-diagonal across zero-, one- and multi-particle sectors. Assuming at most one particle A is produced, the von Neumann entropy of A can be written in terms of scattering probabilities and differential cross sections.
In briefMerging error checking and privacy amplification in quantum key distribution
SummaryQuantum key distribution protocols usually perform error verification and privacy amplification as two separate steps, each applying a hash function to classical data. This paper studies a single hashing step that announces and compares some output bits for error verification and keeps the rest as the final key. The authors show that under certain simple conditions, the security of this combined protocol follows from a slightly modified secrecy notion for the protocol without the final error-verification step. For most entropy-based proofs, the key rate is unchanged and only the security parameter needs a minor adjustment.
In briefMillisecond phonon lifetimes in lithium niobate nanoresonators, limited by external coupling rather than material loss
SummaryThis paper examines why phonon lifetimes in gigahertz piezoelectric nanomechanical resonators have been limited to microseconds. The authors combine chemical restoration of ion-milled surfaces with shadow-mask metallization to fabricate lithium niobate phononic crystal resonators. At 3.6 GHz and millikelvin temperatures, they infer internal phonon lifetimes up to 21 milliseconds. By varying electrode spacing, they show the measured lifetime is set by coupling to the measurement circuit rather than material loss.
WhyChemical restoration and shadow-mask metallization push lithium niobate phonon lifetimes to milliseconds, with electrode-spacing experiments separating external coupling from material loss.
Oliver A. Hitchcock, Matthew P. Maksymowych, Kaveh Pezeshki et al.
In briefWhy forecasting chaotic quantum many-body dynamics can be surprisingly simple
SummaryThis paper treats quantum many-body dynamics as a time-series forecasting problem, using a few-qubit PXP chain that can be realized with Rydberg-atom arrays as a benchmark. By varying the initial state, the system spans regimes from ergodic behavior to quantum many-body scarring. A nonlinear Transformer predicts well in the more ergodic regime but degrades as the state nears the scarred limit, while the simple linear model DLinear stays accurate across all initial states. The results suggest complex quantum evolution can be forecast through a simple linear mapping from past to future observables.
Francesco Perciavalle, Agostino Gallo, Francesco Plastina et al.
In briefQuantum non-Markovian response spectra: probing hidden environmental memory through system-only observations
SummaryQuantum systems are rarely isolated, and hidden environments can retain information and return it later, causing correlated errors. The paper introduces a response matrix that uses only controlled interventions and observations of the system to reveal the temporal structure of this environmental memory. Its spectrum characterizes how memory is structured and persists under specified resets, and suitable combinations of its entries can rule out certain classical memory models. The framework is characterized in an interacting spin process and certifies quantum memory in a two-qubit benchmark.
Aaron Sander, John F. Kam, Gauthameshwar S. et al.
In briefSpin chains with a continuous exponential symmetry and their topological phases
SummaryThe paper studies symmetric gapped ground states of one-dimensional spin-S chains with a U(1)⋊Z2 exponential symmetry, where the U(1) charge unit decays exponentially along the chain. The simplest model is a q-deformation of the XXZ Haldane spin chain. Topology depends on S and q: for integer S with odd q a Haldane-type symmetry-protected topological phase appears, while for even q none does. For half-odd-integer S with even q, the authors identify a new 'half-SPT' phase whose edge mode exists on only one edge.
In briefEfficient Estimation of Logical Sensitivities via Fault Counting
SummaryThis paper proposes a differentiable estimator that recovers how the logical error rate responds to each physical noise parameter from a single Monte Carlo dataset at one noise setting, instead of multiple finite-difference runs. In surface code simulations with circuit-level noise, it matches finite differences while needing one to two orders of magnitude fewer shots for the same variance. The authors apply it to error budgets, qubit-level sensitivities, and effective code distance, and use the sensitivities with Newton root finding to trace threshold contours.
In briefBlack hole radial geometry encoded in a quantum spin chain
SummaryThe paper shows that the radial equation for a scalar field in an anti-de Sitter black brane can be encoded in an exactly solvable semi-infinite spin chain, independent of dimension. The radial coordinate appears as lattice quasimomentum, and spin coherence encodes the geometry. The spectrum and response recover gravitational observables and yield exact screening-mass sum rules. A finite chain is implemented on a superconducting processor.
In briefStandard estimators fail at measured error rates: evidence-based uncertainty for quantum resource estimation
SummaryThe paper argues that estimates of quantum resources for fault-tolerant algorithms are usually single numbers that ignore uncertainty in hardware parameters and disagreement between cost models. The authors propagate evidence-based priors over hardware parameters through five surface-code cost models, using a pre-registered evaluation protocol, and apply it to RSA-2048, ECC-256, AES-256, and a Hubbard-model simulation. Using measured error rates widens the physical-qubit interval roughly fortyfold and raises its median about 4.6 times above the conventional estimate. Two third-party tools exceed their representable range for 80 to 100 percent of the evidence-based parameter space, so these workloads cannot be represented at measured error rates.
WhyApplies evidence-based uncertainty propagation to five surface-code cost models and shows standard tools fail at measured error rates; pre-registered evaluation with multiple workloads gives solid evi
In briefOptimized discrete Wigner representations for detecting non-stabilizerness in qubit systems
SummaryThe paper argues that the lack of a stabilizer-specific phase-space description for qubits comes from using a fixed, state-independent discrete Wigner map. By optimizing the phase-space representation for each state, the authors show that every qubit stabilizer state has an exact delta-function Wigner representation. They define a non-stabilizer witness from the optimized Wigner function at the origin, which stabilizer states saturate for pure states.
In briefErasure qubit protected from low-frequency noise via low-frequency charge driving
SummaryThis paper studies a dynamically protected erasure qubit built on superconducting Kerr oscillators driven by sub-GHz charge signals. The authors find that the oscillator's frequency sensitivity to charge drive grows quadratically with drive frequency and amplitude, making it weakly sensitive to 1/f charge noise while strongly coupled near 1 GHz. They demonstrate 82 ns cooling and reset, and logical control and dynamical protection of a dual-rail qubit with nearly fourfold erasure bias, achieving an error per Clifford of 5.6×10⁻⁴.
WhyExploits frequency dependence of charge sensitivity for fast interactions and 1/f noise protection, with experimental demonstration on dual-rail erasure qubits, though impact remains platform-specific
In briefChecking pulse-level quantum programs against device limits at design time
SummaryThis paper presents qconform, an offline, deterministic checker that decides whether a pulse-level quantum control program can run on a device, based on a versioned capability descriptor for that device. The authors ran differential tests against the QICK and Qblox toolchains on four device configurations, covering 1263 program runs of 969 distinct programs. The checker accepted no program that a toolchain refuses. The evaluation also found and fixed ten defects in the checker and its descriptors.
In briefCV-QKD using noisy coherent states and displacement receivers
SummaryThe paper turns a near-optimal binary coherent-state discrimination method, which uses a displacement followed by photon detection, into a full continuous-variable quantum key distribution protocol. It uses a binary alphabet of noisy coherent states and a displacement receiver, and it proposes a way to estimate channel parameters with the same receiver. The security is analyzed against collective entangling-cloner attacks in the asymptotic regime, and the protocol can generate keys under noisy preparation and imperfect detection, in some regimes outperforming homodyne detection.
L. F. Melo, J. J. Lacerda, John A. Mora Rodr\'iguez et al.
In briefSimulating the Heisenberg XXZ model on a commercial Rydberg atom array
SummaryThe paper implements a domain wall encoding method on a commercially accessible Rydberg atom quantum device, enabling one-dimensional spin chains to be simulated using only Ising-type Hamiltonians. The authors simulate the time evolution of the Heisenberg XXZ model at several anisotropies. They probe its XY and Ising phases and reproduce qualitative behavior of several one- and two-body observables in both phases and at the critical point.
In briefMeasurement-Efficient Differentiable Quantum Architecture Search for Combinatorial Optimization
SummaryDifferentiable quantum architecture search automates quantum circuit design, but its many circuit measurements make hardware execution costly. This paper shows that for a broad class of combinatorial optimization problems and common rotational gate parameterizations, measurements can be substantially reduced without changing the optimization objective. The method is validated on 3-SAT and MaxCut benchmarks, cutting gradient measurement cost by about 39 to 41% with negligible classical post-processing overhead.
Lukas Thei{\ss}inger, Thore Gerlach, Christian Bauckhage
In briefBinary optimization of measurement groups to cut quantum energy estimation costs
SummaryThis paper formulates the grouping of commuting Pauli measurements for quantum energy estimation as a binary clique-selection problem solved with mixed-integer linear programming (MILP). On molecular Hamiltonians, the MILP groupings reduce the non-overlapping measurement requirement by 51.8% on average compared with sorted insertion. Initializing iterative coefficient splitting with these groups gives a further 24.3% average reduction, and the authors also introduce an overlapping-support method, O-clique, along with results on lattice Hamiltonians.
Isaac L. Huidobro-Meezs, Rodrigo A. Vargas-Hern\'andez
In briefPhysics-aligned machine learning for molecular electronic structure that generalizes to larger molecules
SummaryThe paper proposes electronic ground-state descriptor models (GSMs) that sit between machine-learned interatomic potentials and Kohn-Sham density functional theory in computational cost. The authors align learning objectives and architectures with KS-DFT's governing equations by enforcing physical constraints. In size-extrapolation tests from QM9 to QM40, they report large reductions in energy and force errors relative to prior density GSMs, and transfer to the reactive-chemistry dataset Transition1x reportedly reaches energy errors below chemical accuracy.
Eike S. Eberhard, Xaver Kainz, Viktor Kotsev et al.
In briefShot-noise-beating mid-infrared photothermal measurement using twin beams
SummaryThis paper presents mid-infrared photothermal measurement, which detects refractive-index changes from mid-infrared absorption using visible or near-infrared light. The authors use twin beams from seeded optical parametric amplification with balanced detection, suppressing noise more than 5 dB below the coherent-beam shot-noise level. In a proof-of-principle experiment on liquid water, the method achieved an SNR improvement of over 2 dB compared with balanced coherent-beam detection.
In briefGaussian Scrooge ensemble from deep thermalization in free fermions
SummaryThis paper studies how universal ensembles of pure states emerge in a subsystem of a free-fermion chain after measuring its environment, a phenomenon called deep thermalization. The local measurements preserve Gaussianity and break particle-number conservation. The authors argue that the outcomes reveal only partial information about local conserved charges, and this information becomes negligible in the thermodynamic limit. They construct a fermionic Gaussian Scrooge ensemble determined by the subsystem's generalized Gibbs ensemble and give numerical evidence that it describes the late-time projected ensemble at finite temperature.
Angelo Russotto, Katja Klobas, Pasquale Calabrese et al.
In briefExact simulation of open quantum networks with strong coupling via tensor network path integrals
SummaryThe paper proposes CTEMPO, a tensor network scheme that keeps bond dimensions small for single-site non-Markovian problems with strong system-environment coupling. It extends this to simulating quantum networks by time-evolving a comb tensor network. On the spin-boson model, the single-site method reportedly improves on established methods by orders of magnitude. The authors also apply the multi-site approach to the 7-site Fenna-Matthews-Olson complex and to superradiance in up to 20 quantum dots.
WhyProposes CTEMPO with small bond dimensions, extended to networks, validated on spin-boson and FMO benchmarks; strong evidence, notable subfield impact.
In briefQuSema: Using Quantum-Aware AI Agents to Find Silent Bugs in Quantum Software Libraries
SummaryThis paper presents QuSema, an autonomous agent that tests quantum libraries for silent bugs, meaning errors that produce wrong results without crashing. It uses quantum semantics and API documentation as a source-level oracle to check whether valid inputs can yield invalid outputs. On a benchmark of 20 historical silent bugs in Qiskit and PennyLane, it located more bugs on average than Claude Code and Codex, and it found 40 previously unknown bugs confirmed by developers, 30 of them silent.
In briefModular Benchmark of Optimizers for Parameterized Quantum Circuits
SummaryThis paper presents a modular benchmarking framework that separates quantum search-direction estimation from classical parameter-update rules, so their interactions can be studied independently. It compares gradient-based, stochastic, and derivative-free optimizers on QAOA, quantum machine learning classification tasks, and VQE for molecular hydrogen, using both simulators and a 156-qubit processor. The authors note that the hardware runs do not establish an optimizer ranking or isolate the causal effect of device noise.
In briefAn explicit counterexample to Tsirelson's problem using a linear system game
SummaryThe paper builds an explicit binary linear system game that separates the closure of finite-dimensional quantum correlations, C_qa, from commuting operator correlations, C_qc. The game can be won perfectly with a commuting operator strategy, while every C_qa correlation succeeds with probability strictly below 1. This gives a concrete counterexample to the approximation form of Tsirelson's problem. The authors also compute the classical value exactly and formalize the system and separation in Lean 4 with Mathlib.
WhyConstructs an explicit linear system game separating quantum approximation correlations from commuting operator correlations, a concrete counterexample to Tsirelson's problem. Evidence rests on the de
In briefProtocolMatch: A Framework for Choosing Forecasting Models That Depend on Evaluation Protocol
SummaryThe paper introduces ProtocolMatch, an evaluation framework that compares models at matched compute, selects them on validation data, and preserves failure cases for scientific dynamics forecasting. Tested on driven quantum-spin dynamics with recurrent, attention, and linear predictors across three generated datasets, it finds that model rankings shift with training set size, history and closed-loop settings, and system size. Physical penalties did not reliably reduce prediction error, and prediction intervals lost most coverage after a driving-frequency shift. The authors argue that model selection should report accuracy, physical validity, and shifted-distribution reliability separately.
In briefWork Extraction Dynamics in Multi-Atom Systems: The Role of Correlations and Relative Entropy
SummaryThis paper studies how quantum ergotropy, the extractable work, changes over time in a system of two-level atoms interacting with a quantized field. It analyzes the roles of multipartite quantum mutual information and quantum relative entropy. Numerical results for two to five atoms show that maximum ergotropy does not grow linearly with system size, and that zero-ergotropy intervals are modified and suppressed in larger systems.
In briefDesigning Chirped Apodized Crystal Sources for Broadband Quantum Spectroscopy
SummaryThis paper reports the development and characterization of chirped, apodized KTP crystals designed specifically for quantum spectroscopy. The source shows unprecedented bandwidth in a highly non-degenerate configuration. The authors state it could support quantum spectroscopy over broad spectral ranges and enable significantly shorter measurement times.
Linda Sansoni, Max Widarsson, Patrick Mutter et al.
