In briefThermal noise boosts photon-catalysis phase estimation in squeezed thermal interferometry
SummaryThis paper studies phase estimation in a Mach-Zehnder interferometer using squeezed thermal light as the Gaussian resource with difference-intensity detection. It compares photon subtraction, addition and catalysis operations on the heralded state to evaluate phase sensitivity. Single-photon catalysis becomes useful above a thermal occupation of about 0.2 and is the best success-weighted operation over the full thermal range. Quantum Fisher information results show thermal noise enhances phase information, and a loss analysis confirms the effect is robust.
In briefReinforcement learning for controlling solid-state quantum bits
SummaryThis paper studies how model-free reinforcement learning can design pulse-level quantum control strategies for NISQ-era devices. The setup is indirect control of NV nuclear spins in a spin bath, with the bath modeled by the generalized cluster correlation expansion including magnetic impurities and intra-bath interactions. Through simulations, the authors evaluate RL agents under minimally observable environments with noisy, sparse rewards and compare them with other state-of-the-art black-box optimization tools.
In briefComplete classification of perfect state transfer in Grover walks on abelian Cayley graphs
SummaryThis paper completely classifies which connected Cayley graphs over finite abelian groups exhibit perfect state transfer in Grover walks. The authors show that any Cayley graph over a finite group with perfect state transfer has valency 1, 2, or 4. For valency 4, they characterize the abelian case by the set S = {a, -a, a+h, -a+h} with h of order 2 and 4 not dividing |Γ|/2. Together with the valency-one and valency-two cases, all pairs (Γ, S) are explicitly determined, extending earlier results on circulant graphs of valency at most four.
In briefMinimum-length physics for laser light and a proposed two-color vacuum timing experiment
SummaryThis paper studies how coherent light pulses travel under a generalized uncertainty principle, which assumes a minimum measurable length. It shows that a macroscopic multimode pulse's group velocity equals the single-photon value at the central wavelength, times the classical Gaussian-beam factor, independent of total pulse energy. Building on this, the authors propose a two-color vacuum chronoscopy experiment using a petawatt laser and high-harmonic generation, varying beam waist to separate Planck-scale effects from classical diffraction delays.
In briefJianmu-Q: AI Agents Linking Quantum Field Theory to Quantum Hardware
SummaryThis paper introduces Jianmu-Q, a multi-agent framework that converts natural-language descriptions of quantum field theories into quantum simulation workflows. Specialized agents handle theoretical formulation, lattice discretization, qubit mapping, and circuit design, with physical consistency checks and independent numerical calculations used for validation. The framework is tested on the (1+1)-dimensional Nambu–Jona-Lasinio and Schwinger models and the (2+1)-dimensional Z2 gauge theory. Noiseless circuit simulations converge toward independent exact-diagonalization and tensor-network results, and selected circuits were run on IBM quantum hardware.
In briefZZ-Free bSWAP Gates Under Strong Driving in Far-Detuned Transmons
SummaryThis paper studies fixed-frequency superconducting qubits operating with far-detuned frequencies, where a single strong drive produces a large XX-YY interaction and a weak second tone tunes the accompanying ZZ phase. The authors realize a 150-ns √bSWAP gate that can prepare a maximally entangled state directly from the ground state. Compared with cross resonance on the same qubit pair at matched drive strengths, the √bSWAP gives a SWAP operation 33% shorter with higher IRB fidelity.
Akinori Machino, Kazuhisa Ogawa, Yuto Ikegai et al.
In briefCharging a quantum battery faster using spin coupling and noisy environments
SummaryThis paper studies a two-spin quantum battery subject to independent local random telegraph noise, with and without Dzyaloshinskii-Moriya (DM) coupling. Using exact Burshtein-Kubo equations, the authors find that the DM-assisted slow switching protocol gives a superadditive transient ergotropy gain and near-maximal ergotropy in the initial charging stage. In the Ising model, the enhancement is nonmonotonic in DM strength, with intermediate coupling performing best.
In briefOpinion bubbles in the quantum Buridan's ass problem
SummaryThis paper studies the quantum Buridan's ass problem, a network whose node transitions follow quantum mechanics. The authors treat an initially occupied set of coupled nodes as an opinion bubble and examine transitions between it and the rest of the network, as a quantum analogy to changing opinion. They find that quantum interference plays a crucial role in the dynamics, along with the symmetry of the nodes and their connections. Unlike the classical paradox, where symmetry freezes the ass into indecision, the quantum case shows far richer dynamics.
Leonid Chaichenets, Jaroslav Novotn\'y, Igor Jex et al.
In briefBeyond Visibility: Resolving Temporal Structure in Continuous-Wave Four-Photon Hong-Ou-Mandel Interference
SummaryThis paper studies four-photon Hong-Ou-Mandel interference using continuous-wave sources. The authors generate long-coherence photon pairs in microring resonators and record all four photon arrival times with ultrafast time-resolved detection. They map relative delays onto a three-dimensional coincidence histogram relative to a trigger detection, preserving arrival-time correlations usually discarded. The method also allows post-processed optimization of coincidence windows to trade off visibility against count rate.
Sabine H\"aussler, Meritxell Cabrejo-Ponce, Ana Maria Torres et al.
In briefCompact many-body wave functions via selected stabilizer subspaces
SummarySelected configuration interaction (SCI) solves many-body problems by diagonalizing the Hamiltonian in a sparse, iteratively chosen set of configurations, but loses compactness when the target state spreads over many configurations. This paper proposes selected stabilizer interaction (SSI), a fully classical method inspired by SCI that builds a nonorthogonal subspace from a systematically expandable set of stabilizer states. Benchmarks on the transverse-field Ising, XXZ Heisenberg, and Hubbard models against exact diagonalization and equivalent SCI schemes show strong compression in the spin models, with the advantage growing with system size. The authors present this as a proof of principle and note challenges in symmetry handling and exploring the stabilizer space.
In briefHybrid quantum-classical optimization for signal control that gives priority to emergency vehicles
SummaryThis paper studies how to optimize traffic signals that give emergency vehicles priority, using a finite-horizon QUBO model on a three-intersection corridor. It compares exact QUBO optimization, QAOA, CVaR-QAOA, a quantum machine-learning policy, fixed-time control, and max-pressure control across three traffic-demand regimes. Optimization-based controllers reliably moved emergency vehicles in all tested regimes, while max-pressure produced lower ordinary-traffic delay. CVaR-QAOA reproduced the exact QUBO solution more accurately than standard QAOA.
In briefStandard Model generation as a quantum error-correcting code
SummaryThe paper shows that one generation of Standard Model fermions, together with a right-handed neutrino, can be labelled by five bits, a structure linked to anomaly cancellation. The authors argue that anomaly cancellation corresponds to the defining property of triorthogonal codes, which underlie magic state distillation. Removing the right-handed neutrino yields a [[15,1,3]]_p code for every prime p, including the Bravyi-Kitaev code at p = 2. For qutrits, it is the unique shortest distance-three triorthogonal code.
In briefSuperconducting circuits contain defects that last over one second
SummaryThe paper uses flux-tunable transmon qubits and wide-band quantum-jump correlation spectroscopy to identify microscopic two-level systems that outlive the qubit by orders of magnitude. Pump-probe measurements reveal intrinsic lifetimes exceeding one second. Their weak kilohertz-range coupling to the qubit is consistent with positional defects in the standard tunneling model, though other origins such as magnetic spin impurities or trapped quasiparticles are not excluded.
Mathieu F\'echant, Phillip Lenhard, Nicolas Gosling et al.
In briefThree-axis active magnetic compensation to suppress Zeeman decoherence in cold-atom quantum memories
SummaryCold-atom quantum memories need a near-zero magnetic field to limit Zeeman decoherence. This paper proposes a three-axis active compensation scheme that combines spatial feedforward matrix decoupling with a hysteresis PI controller, validated through hardware ablation, stability tests and field mapping. During a 714.4 s steady-hold stage, the central residual field had a mean of 0.112 microtesla and a maximum of 0.224 microtesla.
In briefHankel Submatrices with Bounded Word Length: Dimension Tradeoffs in Probabilistic and Quantum Automata
SummaryThis paper characterizes the largest side length of fixed-margin square Hankel submatrices produced by probabilistic and quantum linear automata with dimension d, when prefix and suffix word lengths are bounded by T. For fixed margins, the log side length is Θ(min{d²,T}) for CPTP quantum realizations and Θ(min{d,T}) for probabilistic automata. At zero false-acceptance tolerance, all four model classes give Θ(min{ln d,T}). The results establish sharp dimension–word-length tradeoffs for bounded-word Hankel tables.
WhyGives tight Θ-order bounds on Hankel submatrix size for probabilistic and quantum automata, with complete proofs, but narrow applicability.
In briefA quantum-compatible conjugate gradient method for preparing ground states
SummaryThe paper proposes quantum conjugate gradient descent (QCGD), a quantum-compatible version of Riemannian conjugate gradient that combines the steepest descent direction with information from earlier search directions. In numerical experiments, it converges faster per iteration than the baseline DB-QITE algorithm. However, each update step has greater circuit depth, creating a trade-off between convergence rate and implementation complexity.
Andres Arcia-Lopez, Marek Gluza, Yudai Suzuki et al.
In briefSame Spectrum, Harder Training: Phase-Locking in Single- vs Multi-Qubit Re-Uploading Circuits
SummaryThe paper compares a single-qubit serial re-uploading quantum circuit with a two-qubit control circuit under the same encoding budget. It argues that the serial circuit's parameters all move the quantum state along one complex direction at each input, a effect called phase-locking, leaving at most 2E+2 tangent directions versus at least 6E+6 for the control. In experiments, the serial circuit reached the optimum in fewer of 100 runs than the control under first-order methods, with mixed held-out error results on real-world series.
Michael Poppel, David Bucher, Maximilian Zorn et al.
In briefLocal finite-rate semi-device-independent randomness expansion via quantum contextuality
SummaryThis paper shows that randomness expansion can be performed in a semi-device-independent way using contextual correlations held by only one localized party, rather than two non-communicating parties sharing Bell nonlocal correlations. The authors propose a template protocol applicable to arbitrary N-cycle non-contextuality inequalities and derive its rate under finite statistics. The template also accommodates non-ideal statistics lacking perfect commutativity and repeatability, and the authors note that several recent loophole-free non-contextuality experiments can be adapted for secure randomness expansion.
In briefProbing a Deconfined Quantum Critical Point on a Digital Quantum Computer
SummaryThis paper studies a deconfined quantum critical point (DQCP), a continuous transition between a z-polarized ferromagnet and a valence-bond solid in a frustrated one-dimensional spin-1/2 chain. The ground states are prepared on a superconducting quantum processor using ancilla-projected quantum imaginary-time evolution, with tensor-network optimization compressing them into shallow circuits. The crossing of the order parameters on hardware locates the transition, and its finite-size extrapolation lies close to large-scale numerical results in the thermodynamic limit.
In briefRemote GKP entanglement between trapped ions
SummaryThis paper proposes a heralded protocol for creating entanglement between trapped ions that are spatially separated. Gottesman-Kitaev-Preskill qubits encoded in the ions' motional modes are linked to photons through the ions' internal electronic states. The authors numerically analyse the fidelity of the distributed logical Bell pair and report that more than 95% is achievable in realistic conditions.
In briefEntanglement entropy of operators under random fermionic Gaussian dynamics
SummaryThis paper studies the operator-space entanglement entropy (OSEE) of initial product operators evolving under random Gaussian dynamics, modeled by a Brownian SYK$_2$ model. By mapping the problem to domain-wall melting, the authors derive an exact expression for the time-dependent bipartite OSEE in the thermodynamic limit and analytically characterize a Gaussian Page curve at late times. The paper also discusses a possible relation between OSEE and non-Gaussianity measures in the Heisenberg picture.
Pablo Bayona-Pena, Elena Gramellini, Lorenzo Piroli
In briefMeasuring How Quantum Programs Diverge Across Vendor Stacks
SummaryThe paper studies realization divergence, the observable difference in how an identical source quantum circuit is realized and behaves across software stacks from several vendors. It reports hardware experiments on three vendors and four devices, combining pre-registered predictions with labeled exploratory analyses. Results show that placement policy strongly affects two-qubit overhead and execution divergence, and that device rankings can invert across hardware. The authors also note that conclusions depend on metric weighting and device generation.
In briefHarmonic light from carbon-chain molecules grows stronger as the chain gets longer.
SummaryThis study simulates high-order harmonic generation from three linear carbon-chain molecules, C2H2, C6H2 and C10H2, driven by an 800 nm linearly polarized laser, using real-time time-dependent density functional theory. Laser intensities were adjusted so each molecule has the same Keldysh parameter, isolating the effect of molecular size. Longer chains produce higher harmonic yields and extend the cutoff, with C10H2 showing much stronger plateau harmonics than C2H2. Time-frequency and charge-density analyses attribute the enhancement to a larger electron acceleration distance and more frequent multi-center recollision, especially near terminal carbon atoms.
In briefLinks between how 'normal' CO2 isotopologue vibrations are and their Raman band strengths
SummaryThis paper studies how the Raman intensities of the strongest bands change across the full series of carbon dioxide isotopologues at room temperature. The authors introduce a scheme for simulating Raman intensities in linear triatomic molecules, based on the mean polarizability surface computed up to cubic terms, and apply it to all twelve stable species. They report correlations between the degree of normality and Raman transition intensities, mean polarizability derivatives, and variational Raman shift errors, checked against experimental spectra of isotopically enriched samples.
Eduardo Su\'arez, Christian Ostertag-Henning, Miguel Carvajal et al.
In briefNumerical method for rare switching in metastable collective spin systems
SummaryThis paper studies rare switching between long-lived metastable states in driven-dissipative collective spin systems. The authors implement a quantum instanton framework numerically by combining collocation and shooting methods, and apply it to the Lipkin-Meshkov-Glick model with collective dissipation and to cooperative resonance fluorescence with collective and local dissipation. The predicted stationary magnetization, first-order dissipative phase-transition point, and slow relaxation scaling agree with quantum master-equation calculations.
Maciej Chudak, Massimiliano Esposito, Krzysztof Ptaszynski
In briefDiffusion language models framed as open quantum systems
SummaryThe paper shows that the continuous-time forward and reverse dynamics of diffusion language models can be embedded in the framework of open quantum systems. Building on this, it proposes generating text from superpositions of prompts or thoughts (SoP, SoT) and a quantum-inspired remasking rule (QIR) that revises previously generated tokens. Experiments with pretrained models report that superposition-based generation retains information from conditioning components not selected at measurement, and that QIR improves LLM-assessed text quality when applied at an appropriate stage.
Akinori Tanaka, Yukinao Akamatsu, Takuki Tomita et al.
In briefDispersion and exchange forces with surface roughness for signal transmission in a nanoswitch
SummaryThis paper develops a theoretical model of a resistive nanoswitch that switches between two operating modes, accounting for electrostatic, elastic, dispersion (van der Waals and Casimir) and exchange repulsive forces with surface roughness. The equilibrium closest separation between the cantilever tip and ground plate is calculated as a function of actuation voltage, and the device cycles on and off without pull-in. The authors also compute the tunnel current and show that increasing the amplitude difference between carrier and information signals can make low-frequency amplitude distortions sufficiently small.
Galina L. Klimchitskaya, Alexander S. Korotkov, Vera V. Loboda et al.
In briefFinite-size geometric crossover toward fragmentation of potential energy shells in the quantum random energy model
SummaryThis paper studies the connectivity on the Fock-space hypercube of a potential energy shell, defined by energy variance, in the quantum random energy model. As the transverse field varies, the energy-selected subgraph shows a finite-size geometric crossover between fully connected and fragmented regimes. The authors also report a correlation between shell connectivity and the finite-size fractal dimension of eigenstates extracted from participation entropy.
In briefFast 3D parallel rearrangement of neutral atom arrays using combined elementary holograms
SummaryThe paper presents a fast, parallel method for three-dimensionally rearranging stochastically loaded neutral atoms into defect-free arrays. Holograms are computed analytically by summing the complex fields of individual traps, with trap phases ramped linearly during motion and continuous trap positions for high accuracy. Using one GPU, each intermediate hologram takes 0.284 ms, faster than the 0.696 ms SLM refresh time, and a 100-atom array is assembled in under 20 ms with 99.31(12)% per-atom survival. The method also demonstrates axial atom transport, including a leapfrog protocol.
In briefSpeeding up quantum cooling using classical noise
SummaryThis paper studies how Markovian classical noise affects the cooling, or freezing, of a quantum system. It notes that under certain hypotheses anomalous thermalization, such as the Mpemba effect, can occur, where a hotter initial state cools faster. The authors show that classical noise during freezing can accelerate the approach to the zero-temperature thermal state, a counterintuitive result.
In briefPractical Fidelity Estimation for Noisy Quantum Devices Using Direct and Survival-Probability-Based Protocols
SummaryThis paper presents two methods for estimating quantum state fidelity without full tomographic reconstruction, aimed at noisy intermediate-scale quantum devices. The first, direct fidelity estimation, uses importance-sampled Pauli measurements, while the second reconstructs fidelity from moments of a noisy evolution and its approximate inverse and is robust to SPAM errors. Numerical tests under depolarizing, dephasing, and amplitude-damping channels show that attenuation-aware correction improves robustness to measurement noise, and that an L2-optimized expansion outperforms naive Taylor truncation in high-noise regimes.
Ahmed Aman, Kaustav Chatterjee, Younes Javanmard et al.
In briefTime-optimal robust quantum control when error classes are partly known
SummaryThe paper studies how quickly a quantum control can cancel, to first order, all static errors within a known class whose strength is unknown. A time-independent Hamiltonian can do this exactly when all errors in the class have zero diagonal in a common basis, and an equally spaced energy ladder in that basis achieves it in one period. For classes such as unknown level shifts and unknown local errors on two subsystems, this constant control is shown to be optimal. For completely unknown errors, a time-dependent control is required, and knowing the error class shortens the minimum time by a factor of √(d²+1).
WhyProves tight time-optimal bounds for robust control under partially known error classes, showing constant control is optimal for several natural classes, a useful advance for the subfield.
In briefLearning sparse SNAP gate structures to cut the cost of qudit state preparation
SummaryThis paper studies preparing bosonic qudit states using SNAP-Displacement sequences, where each gate's phases must be chosen and placed. The authors propose a differentiable masking method that learns which phases are needed along with their values, using a sparsity penalty and temperature annealing. They compare it with fully parameterized and predefined sparse ansatzes on Fourier and Haar-random target states. The results show competitive or improved fidelity-resource trade-offs, though the advantage depends on the target state and qudit dimension.
Marco Pennazio, Leonardo Ignazio Pagliochini, Riccardo Pellini et al.
In briefBounds on boson number near the quantum Rabi phase transition at finite frequency
SummaryThis paper studies the quantum Rabi model, a basic quantum optics model of a two-level system coupled to a bosonic mode including counter-rotating terms. The authors derive rigorous upper and lower bounds on the ground-state energy and boson occupation number across the quantum phase transition at finite frequency ratios. Exact numerical diagonalization shows the bounds are tight, and the resulting inequalities offer predictions testable in trapped-ion quantum simulators.
Razmik G. Unanyan, Michael Fleischhauer, and Nikolay V. Vitanov
In briefHow late-time photons "record" primordial curvature perturbations
SummaryThe paper addresses the gap between primordial quantum fluctuations in inflationary cosmology and the classical cosmic history seen in observations. It introduces Cosmological Classical Branches (CCBs), using the quantum-information notion of physical records, in which distinct late-time matter components carry compatible records of the same primordial configuration. This gives each branch a common classical label at finite resolution while the global superposition persists. The Letter summarizes a framework developed in a referenced earlier work.
In briefLimits of adaptive two-way linear postprocessing in six-state quantum key distribution
SummaryThis paper studies how far adaptive linear postprocessing with two-way classical communication can improve noise tolerance in six-state quantum key distribution. The authors construct a single compatible collective attack that defeats a broad class of linear postprocessing. They prove that Eve's optimal probability of recovering any linear function of Alice's raw key is at least Bob's, so reliable agreement on a linear final key is incompatible with secrecy.
WhyConstructs one compatible collective attack showing adaptive linear postprocessing limits in six-state QKD, with self-contained proofs supporting a strong theoretical claim.
In briefControlling quantum states and detection statistics on a photonic chip via periodic conditional measurements
SummaryThis paper studies periodic conditional measurements, in which coherent evolution is repeatedly interrupted by local measurements. The authors use bright and dark modes of the conditional evolution operator to describe both post-selected final states along no-detection trajectories and detection-event statistics. Using a three-dimensional photonic chip, they demonstrate on-demand state engineering, including protection of topological corner states and excited-state shelving, and observe a transition in total detection probability on an effectively infinite one-dimensional lattice.
In briefCasimir force between graphene-covered suspended membranes
SummaryThis paper studies the Casimir force on thin suspended dielectric membranes covered with graphene, using both theory and experiments. The authors show that replacing the solid substrate under graphene with a thin membrane partially restores a strong temperature dependence of the force. Theory predicts pressure changes of up to 54% for silicon oxide and 40% for silicon nitride at about 500 nm separation when heated from 10 K to 294 K. Room-temperature measurements on silicon nitride matched the predicted force gradient, but the temperature dependence could not be resolved.
In briefPrecise gap control for 3D-integrated superconducting quantum circuits via asymmetric indium bonding
SummaryThis paper presents a flip-chip bonding method for 3D-integrated superconducting quantum chips, using thick indium bumps on the control chip and thin indium pads on the qubit chip. Electrical and microwave tests show low-loss indium interconnects and transmon qubits with T1 of (159±14) μs and T2,echo of (294±42) μs, comparable to planar devices. A multi-angle compression step reduces the average post-bonding tilt to (20±12) μrad.
In briefSwitching Between Time-Bin and Energy-Time Photon Entanglement Using Homodyned Resonance Fluorescence
SummaryThis paper theoretically studies photon entanglement produced by homodyning resonance fluorescence from a two-level emitter with a phase-locked laser field under weak excitation. Increasing the relative laser amplitude causes entanglement to crossover from time-bin to energy-time and back to time-bin. The energy-time regime appears near amplitude matching, where destructive interference suppresses the single-photon component. The entanglement is characterized by violations of the CHSH Bell inequality in a Franson interferometer.
In briefOptimizing multipartite entanglement manipulation in quantum networks
SummaryThis paper studies how to manipulate genuine multipartite entanglement (GME) in quantum networks so that distant nodes can share long-range GME on demand. The authors present a general framework that groups networking operations into three primitives, inflation, deflation, and purification, which can be combined into multipartite quantum repeater protocols. They apply the framework to networks of Dür-Cirac-Tarrach states, showing the achievable end-to-end GME and accessible entanglement classes.
Shashank Shekhar, David Elkouss, Siddhartha Santra
In briefAxion-dilaton wormhole partition function away from the BPS limit
SummaryThe paper computes the leading finite-radius correction to the two-boundary response of a charged axion wormhole near its BPS endpoint. Using exact radial matching in an axion-dilaton system, it identifies the range where the quadratic approximation holds and the first angular resonance. The inverse response is then averaged over the chiral Gaussian ensemble describing the BPS endpoint.
In briefParallel repetition bound for entangled games via conditional relative entropy
SummaryThe paper proves a cubic parallel-repetition bound for finite two-player entangled games using quantum conditional relative entropy. If a game has entangled value 1-ε with ε<1, its n-fold repetition value is at most exp[-Ω(ε³n/L)], where L is the maximum of 1 and the log of the product of the answer alphabet sizes. A threshold bound is also given, showing winning at least a 1-ε+δ fraction has probability at most exp[-Ω(δ⁵n/(ε²L))]. The result recovers Song's cubic rate.
WhyUses quantum conditional relative entropy to prove a cubic parallel-repetition bound for entangled games, recovering Song's rate and adding a threshold bound with a complete proof.
In briefWhen Do Quantum Views Help? Complementary Classical-Quantum Fusion Explained
SummaryThe paper proposes a multi-view framework that keeps the full classical input and adds simulated quantum features, combining them via concatenation or attention. Controlled experiments show fusion helps when classical and quantum representations capture complementary predictive relationships. Classical replacement and masking controls remain competitive, and classical baselines lead on some benchmarks such as Digits and Covertype.
Azadeh Alavi, Fatemeh Kouchmeshki, Hossein Akhoundi et al.
In briefStitched surface codes: a fault-tolerant quantum memory linking patches via GHZ corridors
SummaryThis paper proposes a fault-tolerant quantum-memory design that combines four rotated surface-code patches with an outer [[4,2,2]] code, encoding two logical qubits at distance 2t on a square lattice. Outer-code stabilizers are measured through verified GHZ ancillae in narrow corridors between patches, and decoded in a single matching pass. On a 105-qubit superconducting processor, raising the code distance from 4 to 6 reduced the block logical error rate by a factor of 1.52.
Mohamed A. Shalby, Alexandre Bourassa, Amir Karamlou et al.
In briefRegret lower bound for quantum state learning with log loss and related tomography and confidence results
SummaryThis paper studies online learning of quantum states under logarithmic loss, a quantum analogue of online portfolio selection with uses in quantum state tomography. The authors prove a regret lower bound showing that Exponential Weights and Tseng et al. (2025)'s algorithm have optimal polynomial dependence on dimension. The proof yields a tomography algorithm that is sample optimal up to polylogarithmic factors, and a method for confidence sequences for matrix-valued random variable means, which a numerical experiment shows is competitive.
In briefLocalized states of a non-Hermitian Schrödinger equation with a complex 1/r² potential, linking Efimov states and bipolarons
SummaryThis paper studies a non-Hermitian Schrödinger equation with a complex 1/r² potential, linking Hermitian few-body and non-Hermitian many-body physics. A real attractive potential describes Efimov bound states, while a purely imaginary potential describes the medium-induced interaction between two heavy polarons at large separations. The authors solve the equation analytically and compare the results with numerical and WKB solutions. As the imaginary part grows, Efimov bound-state energies acquire imaginary parts and rotate counterclockwise in the complex plane, and the ground state connects smoothly to a short-distance bipolaron localized solely by the imaginary interaction.
In briefGravitational-wave detector using photon orbital angular momentum tomography, free of displacement noise
SummaryThe paper proposes detecting gravitational waves by measuring transitions between photon orbital-angular-momentum (OAM) states rather than differential arm lengths, so it is intrinsically free of displacement noise. The authors claim a single V-shaped detector could resolve all seven wave parameters via parametrized quantum state tomography, with a reference state added to boost the signal. They estimate a shot-noise-limited sensitivity of 1.6×10⁻²⁰ Hz⁻¹/² and a broadband response bridging the 0.1–10 Hz mid-frequency gap.
In briefSpeeding up exchange calculations in quantum chemistry on multi-core CPUs with SIMD instructions
SummaryThis paper presents SNK, a seminumerical exchange algorithm based on the sn-LinK method, optimized for modern multi-core CPUs with SIMD vector instructions and implemented in the Q-Chem program. The authors describe vectorization and memory strategies for Boys function and three-center one-electron integral kernels, generated symbolically with SymPy and exported as optimized C++ code. Tests on systems of 100-500 atoms and 2000-10000 basis functions show 1-2 orders of magnitude speedups over traditional four-center exchange. For smaller molecules, SNK surpasses occ-RI-K beyond roughly 50-100 atoms.
In briefSharper upper bounds on zero-error capacity for quantum channels with no-signalling assistance
SummaryThis paper studies zero-error communication and exact simulation of quantum channels aided by quantum no-signalling correlations. The authors introduce a semidefinite programming parameter Γ(K) that is exactly multiplicative under tensor products and upper-bounds the no-signalling zero-error capacity quantity Θ_NS, and it is the tightest among known SDP upper bounds. They also show the simulation lower bound Σ⁻(K) is multiplicative, which implies that neither the single-copy nor the regularized asymptotic simulation cost formula holds in general, answering an open question. For classical-quantum channels, sharp bounds depending on output dimensions are obtained.
In briefAgent-assisted transfer of experimental strategy across quantum processors
SummaryThis paper presents an agent-assisted experimental strategy for running experiments on three-qubit NMR quantum processors. A strategy developed on one processor was applied sequentially to four others, with an agent assessing results and selecting measurements while operators handled instrument preparation. Using three-qubit Grover state preparation as the shared task, all four processors produced outputs dominated by their marked states. The marked-state populations ranged from 0.910 to 0.959, with relative deviations below 3.8% from the ideal value of 0.945.
In briefSingular wave functions and quantum collapse in inverse-power potentials
SummaryThis paper studies quantum collapse under the inverse-power potential U(r)=-U0/(2r^n) within the Schroedinger equation. The stationary wave functions oscillate infinitely often as r approaches zero, and the ground state is absent when collapse occurs. The authors give an asymptotic form of wave functions filling the continuous spectrum at r→0, along with some exact solutions under suitable conditions. As a regularized case, they consider the potential truncated at a finite distance from the center, where the wave functions have finitely many oscillations, the ground state reappears, and the bound-state spectrum becomes discrete.
In briefSimplifying Drone Delivery QUBOs with Classical Feasibility Preprocessing
SummaryThis paper studies drone-delivery mission selection, where candidate missions are first generated and screened classically to remove payload- and energy-infeasible options before building a QUBO model. On synthetic instances, exact enumeration and simulated annealing matched the MILP optimum, while the evaluated QAOA configuration achieved lower feasibility and was limited to small instances. Preprocessing reduced mean QUBO variables by 29.60% and simulated annealing runtime by 31.26%.
Silvie Ill\'esov\'a, Vojt\v{e}ch Nov\'ak, Martin Beseda
In briefmonoprop: Scaling Pauli and Majorana Operator Propagation with Anticommutation Indices
SummaryThe paper presents monoprop, an open-source, massively parallel framework for propagating operators in fermion and qubit bases. It introduces an anticommutation index that quickly finds the operator terms anticommuting with a gate, with per-gate time scaling as O(K/N) under the stated truncation. The authors report near-ideal scaling to 8192 cores while tracking 9.5×10^10 terms, and describe surrogate graphs for fast energy re-estimation and parameter gradients.
In briefClassifying quantum states via conserved operators, a framework for deciding which states can convert into one another
SummaryThe paper proposes a classification framework for quantum states based on conserved operators, which are matrices having a given state as an eigenstate. The authors show these operators can be obtained from the kernel of a quantum covariance matrix and derive general bounds on their number. They apply the approach to multiqubit SLOCC convertibility, giving classifications for four, five, and six qubits.
WhyIntroduces a conserved-operator framework with general bounds and refines 4-qubit SLOCC classes to 29 types, extending exhaustive classification to 5 and 6 qubits.
In briefSemiclassical Wave Dynamics via Wigner Transforms: Quantum Harmonic, Anharmonic and Dirac Oscillators
SummaryThis Master's thesis offers a pedagogical introduction to semiclassical analysis of wave dynamics in single-band and multi-band systems. It studies the quantum harmonic and anharmonic oscillators, and the Dirac oscillator, showing how Wigner function evolution approximates wave packet centers and deformations. It also compares this approach with the Sundaram-Niu formalism and examines the conditions for the semiclassical approximation to be valid.
In briefCasimir Energy Sign Reversal from Finite-Range Quantum Fluid Fluctuations
SummaryThis paper studies the zero-temperature Casimir effect in a three-dimensional Bose fluid whose Bogoliubov spectrum includes finite-range corrections. Besides the universal acoustic term scaling as L^-3, a nonuniversal L^-5 correction can compete with it. The authors show that robust sign changes of the Casimir energy occur near roton softening, and discuss how the resulting force could be probed with a driven nanomechanical resonator.
In briefA Rydberg atom array coupled to a high-cooperativity optical cavity
SummaryThis paper combines Rydberg-state control of atom-atom interactions with optical-cavity control of atom-light interactions. The researchers coupled a Rydberg tweezer array to a high-cooperativity cavity and suppressed the electric fields from cavity length actuation so they did not limit Rydberg coherence. They demonstrated Purcell-enhanced fluorescence readout within 7.5 microseconds and used Rydberg-Rydberg interactions to control collective light scattering into the cavity mode.
WhyCouples Rydberg tweezer arrays to a high-cooperativity cavity, demonstrating Purcell readout and blockade-controlled collective scattering with solid experimental evidence; impact mainly within quantu
In briefA tutorial on quantum scattering and coupled-channel calculations for molecular collisions
SummaryThis tutorial formulates the quantum scattering problem in Dirac notation and builds a practical numerical framework from it. It derives the coupled-channel representation with correct asymptotic behaviour and uses Numerov's method to obtain time-independent wavefunctions. From these, cross sections, the observables measured in scattering experiments, are calculated, progressing from single-channel scattering to collisions of a ground-state atom with a diatomic molecule.
In briefPerfect and peak state transfer experiments on IBM's 156-qubit quantum processor
SummaryResearchers ran continuous-time quantum walks on IBM's 156-qubit Heron superconducting processor, ibm_kingston, without error correction or mitigation. The experiments test perfect state transfer on hypercubes and related graphs, and peak state transfer on the X_n family. The results are compared with moving a state along a qubit path using SWAP gates.
In briefEntanglement in superpositions of zero-energy states, analysed with stabilizer theory
SummaryThe paper studies superpositions of exact zero-energy eigenstates built from entangled dimers in a class of spin Hamiltonians. Using stabilizer theory and polynomial methods, the authors characterize the subspace these superpositions span and find the minimum entanglement entropy between two halves of a periodic spin-1/2 chain. For Hamiltonians generated by a single local Pauli interaction with noncommuting translates, this minimum follows a volume law at fixed interaction range.
In briefDecoding qubit readout errors by syndrome, per error channel
SummaryThe paper introduces the Readout Error Decoder (RED), which turns repeated thresholded binary readout records into an error budget broken down by error channel. Controlled error injections are used to validate how the decoder responds to each channel. RED is demonstrated on fluxonium and transmon qubits across different leakage-lifetime regimes. It identifies the error channels limiting readout performance without requiring dedicated discrimination of non-computational states.
Han Zheng, Vladimir Pe\v{s}i\'c, Nikolay Gusarov et al.
In briefCollective organization behind how strings break into particles
SummaryThe paper uses real-time tensor-network simulations of the massive Schwinger model to study how coherent confining strings turn into hadronic matter. For intermediate masses, over 99% of the electric-field distribution's temporal evolution is governed by a single collective mode whose decay tracks the parent string's survival probability. The extracted survival law shows the same Gaussian dependence as the Lund world-sheet area law.
WhyReal-time tensor-network simulations reveal a single collective mode governing string fragmentation, microscopically supporting the Lund area law with solid evidence, mainly affecting gauge dynamics.
Romuald A. Janik, Maciej A. Nowak, Marek M. Rams et al.
In briefSparse-QITE: a regularized method for finding compact generators for quantum imaginary-time evolution
SummaryQuantum imaginary-time evolution (QITE) is a method for preparing ground states, but its cost rises quickly with the locality parameter D. This paper proposes Sparse-QITE, a regularized formulation that favors compact generators, reducing the number of quantum operations and producing circuits with fewer gates and lower depth. Benchmarks on one-dimensional transverse-field Ising and antiferromagnetic Heisenberg models show a controllable sparsity–accuracy trade-off, with some large cases needing about 1% of standard QITE's gate count while keeping high accuracy.
Rafael G\'omez-Lurbe, Luca Ion, Alberto Acevedo et al.
In briefInfinite magic and the emergence of the continuum: how quantum states' nonstabilizerness (magic) relates to whether lattice systems can give rise to relativistic quantum field theory.
SummaryThe paper quantitatively links the type classification of hyperfinite von Neumann algebras to nonstabilizerness, or magic, a quantum-information resource. It shows that local algebras of stabilizer systems on a lattice contain no type III factors, and for short-range-ordered stabilizer systems they are tensor products of type I or type II hyperfinite factors. For Araki, Woods and Powers hyperfinite factors, the limiting factor is type III exactly when the locally optimized min-relative entropy of magic diverges.
WhyEstablishes a quantitative link between von Neumann algebra type and magic, proving type III factors require divergent min-relative entropy of magic.
Leo Shaposhnik, Lorenzo Leone, Lennart Bittel et al.
In briefBeyond the Lovász number: a continuum of Rényi points in graph spectra
SummaryThe paper studies the entanglement-assisted asymptotic spectrum of graphs introduced by Li and Zuiddam. The authors show that, besides the Lovász number, a family of parameters obtained by interpolating Rényi divergence orders between 0 and 2 also belongs to this spectrum. They prove a conjecture of Duan and Winter on the minimal Holevo capacity of the five-cycle and settle two open links in their inequality chain.
WhyConstructs a continuum of new spectral points and proves the Duan–Winter conjecture with rigorous arguments, though impact is confined to graph theory and quantum information.
In briefThe Magic Barrier: A Universal Feature of Quantum Thermalization
SummaryThis paper studies quantum thermalization under unitary dynamics generated by random matrix Hamiltonians, starting from a pure product state. The authors analytically derive the time evolution of entanglement entropy, antiflatness, capacity of entanglement, and nonlocal trace distance magic. They report that the transient maxima of these quantities occur on the same timescale as the maximum rate of entanglement growth. The authors conclude that this “magic barrier” is a universal feature of random matrix theory thermalization.
Jakob J. Minar, Andreas Sch\"afer, Leonhard Schmotzer et al.
In briefStandard and bipolar rectification of particle and energy flow in tilted, periodically driven quantum systems
SummaryThis paper studies a fermionic system coupled to external baths with a tilted potential and a locally applied time-periodic drive. The drive produces particle and energy currents even without chemical or thermal bias between the baths, and these currents are enhanced at frequencies where avoided crossings occur in the Floquet quasienergy spectrum. A rectification coefficient R quantifies the asymmetry between forward and reverse bias transport, revealing standard and bipolar rectification regimes, with perfect rectification at their boundary. Including interparticle interactions enlarges the bipolar regime and can raise R by many orders of magnitude.
Joaqu\'in Pe\~{n}uela-Parra, Juan Jos\'e Mendoza-Arenas
In briefSchrödinger's unpublished hidden-variable model in the 1944–45 debate on determinism
SummaryThis paper examines a mid-1940s debate in Nature over whether quantum mechanics could be supplemented by deterministic hidden variables. The author shows that Erwin Schrödinger drafted an unpublished response that presented an explicit hidden-variable model, demonstrating that such models could not be ruled out. The paper reconstructs the historical context of this forgotten manuscript and publishes its first transcription and analysis. It argues that the proposal should be read as a proof of principle rather than as evidence of Schrödinger's commitment to determinism.
In briefA constructive method for the n-representability problem, using two-body witnesses and numerical checks
SummaryThe paper builds the full family of two-body witnesses at each level of a weighted sum-of-squares hierarchy. It combines positive-semidefinite Gram matrices with explicit linear nullspace constraints that cancel higher-body contributions. The authors solve the associated semidefinite program and recover the exact ground-state energy of a Richardson Hamiltonian in the three-fermion sector. They also prove that higher-degree particle-number multipliers are unnecessary when the full sum-of-squares generating algebra is used.
In briefAlternating squeezing and photon subtraction to generate optical cat states
SummaryThis paper analyzes an approach that alternates squeezing and photon-number-subtraction operations to generate optical Schrödinger cat states. The authors formulate generation as a fidelity-constrained multibranch optimization, jointly maximizing the total probability of all photon-number outcomes that meet a fidelity threshold. Under experimental limits on squeezing and photon-number resolution, the scheme improves fidelity and success probability over plain photon subtraction.
Johanna Conrad, Fabian Steinlechner, and Ren\'{e} Sondenheimer
In briefQuantum Prepare-and-Measure Scenarios: Concepts and Methods
SummaryThis tutorial introduces quantum prepare-and-measure scenarios, in which a sender prepares a quantum system in one of several states and a receiver measures it. It presents frameworks for constraining and characterising communication, with emphasis on semi-device-independent approaches. The paper also covers modern convex-optimisation techniques for characterising achievable correlations and aims to give newcomers the foundations needed for the current literature.
A. de Oliveira Junior, Santiago Zamora, Jonatan Bohr Brask et al.
In briefBoundary-induced geometric frustration in square Rydberg-atom arrays
SummaryThis paper studies geometric defects in square arrays of Rydberg atoms caused by whether the lattice size is odd or even. Odd-sized lattices form a single striated domain, while even-sized lattices create incompatible domains at the four corners, leaving a defect from the center. The authors tested the preparation protocol on the Aquila quantum processor and compared it with tensor-network simulations.
Marco Rigobello, Luka Pave\v{s}i\'{c}, Daniel Jaschke et al.
In briefLong-lived local coherences at high temperature from frozen domain walls
SummaryThis paper studies why local coherences between distinct charge sectors in spin chains can persist anomalously long at high temperature. Using a Mazur bound derived from quantum-group symmetry, together with locality and a non-undershoot assumption, the authors show the transverse spin autocorrelation can decay at most algebraically, C(t) ≳ 1/t². For the XXZ chain at infinite temperature in the large-anisotropy limit, they explicitly compute C(t) ~ √3/(18√t).
In briefCharge sharpening in monitored quantum circuits survives readout errors and incomplete postselection
SummaryThis paper studies how readout errors affect monitored quantum circuits, making postselection incomplete because trajectories are conditioned on imperfect records. For U(1)-symmetric circuits, the charge-sharpening transition survives at finite error rates, with a finite steady-state plateau replacing vanishing fluctuations. The authors decompose fluctuations into quantum and classical parts to show a detectable signature of the transition that can be located directly from experimental data.
In briefFinite-cutoff geometry in double-scaled SYK: deformations, branes, and complexity
SummaryThis paper develops a finite-cutoff holographic description of the double-scaled SYK model. It constructs deformations of KM and thermofield-double states that reproduce lower-dimensional analogues of TT̄ deformations, and studies the resulting wormhole and AdS/dS geometries in sine dilaton gravity. It also incorporates end-of-the-world branes and evaluates finite-cutoff lengths of constant mean curvature slices, interpreted as boundary complexity.
In briefQuantum partial information decompositions and direct measures of objective information
SummaryThe paper extends classical partial information decomposition to multipartite quantum systems, constructing two quantum PIDs of the Holevo quantity. One is a naive version based on classical minimum mutual information; the other relies on a new operational partial order on quantum ensembles defined by ensemble interconvertibility. The authors apply these decompositions to directly quantify objective information in extended graph states.
In briefLogical CCZ state cultivation: a protocol for preparing logical CCZ states for Toffoli gates in fault-tolerant quantum computing
SummaryThis paper presents an end-to-end protocol for directly preparing logical CCZ states on bivariate tricycle and bicycle (BT/BB) codes, which enable Toffoli gates. It combines the native cup-product CCZ gates of BT codes with two new techniques based on projection-preserving lifts, and transfers the prepared states to high-distance lifted BB memories using decoder-based postselection. Exact simulations under a neutral-atom-motivated circuit-level noise model report joint infidelities of 1.84×10⁻⁵ at 21.94% acceptance and 4.10×10⁻⁷ at 8.89% acceptance. The authors estimate that the smallest factory's spacetime cost per CCZ state is lower than that of T-state cultivation followed by CCZ conversion at the same output infidelity target.
WhyProposes direct logical CCZ preparation on BT/BB codes with projection-preserving lifts and decoder postselection, supported by exact circuit-level simulations showing low infidelity and cost.
In briefMemory costs for synthesizing quantum channels with a protected subsystem
SummaryThe paper studies how much total intermediate dimension is needed to synthesize a quantum channel when a first stage perfectly preserves a subsystem while discarding information a later operation needs. It gives a condition for exact synthesis of a finite family of classically controlled branches and constructs a single processor implementing all of them. For a trine-based family, a four-dimensional quantum carrier achieves exact synthesis under output depolarization, while classical programs of the same dimension have higher error.
In briefQuantum work statistics via entanglement-assisted invariance (envariance)
SummaryThe paper compares the one-time and two-time measurement schemes for quantum work statistics. It argues that the one-time scheme is more consistent with envariance, the fundamental symmetry of entanglement. Work statistics then follow for any driving unitary without an explicit measurement postulate, whereas the two-time scheme does so only for a Haar-measure-zero set of unitaries. This removes ambiguity tied to the energetic cost of measurement.
In briefApproximate quantum error correction recovery via a single QSVT polar transformation
SummaryThis paper implements the Petz recovery map, a near-optimal decoder for approximate quantum error correction, using a single quantum singular value transformation (QSVT) based on the polar structure of its recovery isometry. This avoids the difficult circuit implementation of the inverse square-root normalization. The authors give explicit block-encoding circuits and an extended recovery isometry that works without postselection. They also bound the error using a spectrum-weighted loss and analyze the tradeoff between circuit depth and accuracy for Haar-random encoding and the Hayden-Preskill protocol.
In briefA portable ion-trap quantum device demonstrated in transit
SummaryThe paper reports the first transport of a few thousand ions held in a linear Paul trap over several kilometers. The transports lasted 15 and 45 minutes. The results show that trapped-ion devices can operate while in motion, providing a foundation for mobile quantum devices and for moving exotic ions between experimental facilities.
Sara Alfaro, Dipesh Bhanushali, Sankarshan Choudapurkar et al.
In briefQuantum spectral estimation using multiple initial states with improved error and cost scaling
SummaryThis paper studies estimating quantum spectra and eigenstate properties using multiple initial states and generalized Hadamard tests, building on the QFAMES framework. The authors propose a method that samples evolution times independently for each initial-state pair. Under stated assumptions, when statistical error sets the sampling cost, the total Hamiltonian simulation cost is independent of the number of initial states, up to logarithmic factors. The method also keeps Heisenberg-limited scaling for eigenvalue estimation and reduces multiplicity recovery cost.
In briefUniversal enhancement of the pairing gap in a dilute Fermi superfluid
SummaryThis paper studies a dilute Fermi gas with attractive contact interactions, which at zero temperature forms a superfluid with gapped fermionic excitations. The authors show that the weak-coupling gap can be written as a trans-series expansion in the interaction strength and the asymptotic Gor'kov-Melik-Barkhudarov gap. Corrections from low-energy virtual quasiparticles are linear in the gap, with a leading coefficient nearly 25 times larger than other known coefficients, which substantially increases the gap at intermediate densities. The work emphasizes that systematic corrections are crucial for quantitative predictions.
Silas R. Beane, Roland C. Farrell, Cameron P. Shotwell
In briefProof of a factorised determinant formula for monomial bases in commutative semisimple algebras, via higher-dimensional Young diagrams
SummaryThe paper proves a conjectured factorised formula for the determinant of a monomial basis of a commutative semisimple algebra, built from a degeneracy graph. The proof uses graph analogues of Newton polynomials and induction on the number of layers. The exponents in the determinant are shown to be overlap volumes of higher-dimensional Young diagrams attached to the graph's nodes. The block triangular structure also yields an explicit recursive inversion of the evaluation matrix.
WhyProves the Kemp-Ramgoolam conjectured factorised determinant formula for any number of generators, a strong theoretical result with a narrow scope.
