Papers

For many important optimization problems we are restricted to approximate solutions in practice due to computational complexity. Distinct from the exact optimization setting, approximate optimization admits performance measures beyond whether the optimum is found, with different tradeoffs and complexity. For MaxCut, a near-unity (ordinary) approximation ratio can coexist with near-zero improvement (gain) over a random cut. For the standard encoding, the unconditional classical MaxCut-Gain hardness gap implies that any uniformly efficient quantum or hybrid procedure recovering a fixed positive fraction of the optimal classical gain on every input, with at least inverse-polynomial success probability, would place NP in BQP. Such a procedure is therefore believed impossible under standard assumptions. We broadly address where our worst-case barriers do or do not apply across the quantum algorithm landscape. We prove that the barrier survives quantum random access optimization (QRAO) compression and applies between the classical and relaxed optimal values. For every input, a product state attains the classical optimum. Thus the barrier to reaching the classical threshold does not arise from a need for entanglement. For $d\in\{2,3\}$ variables per qubit, the known decoder transfers encoded energy gain to decoded mean gain by the exact factor $1/d^2$. Combining this identity with MaxCut-Gain hardness gives an operational preparation barrier for QRAO. We also construct hard $n$-qubit families with relative quantum relaxation excess $\Theta(1/n)$, while the maximally mixed state has energy approximation ratio $1-\Theta(1/n)$, zero encoded energy gain, and hence zero decoded mean gain. Our results separate the effects of relaxation tightness and energy approximation from operational accessibility, motivating more comprehensive accounting in benchmarking and performance assessment.

The Simultaneous Authentication of Equals (SAE) protocol, introduced in WPA3, provides robust protection against offline dictionary attacks against a network Pre-Shared Key (PSK) and also protection of session keys from other people knowing that PSK. However, its high computational cost for the Password Element (PE) derivation makes Access Points (APs) vulnerable to CPU exhaustion Denial-of-Service (DoS) attacks. This paper proposes a resilient architectural modification to SAE. First, we introduce an asymmetric resource cost model that offloads the iterative discovery of cryptographic elements to the client, allowing the AP to maintain a fixed computational load during the handshake. To further mitigate brute-force attempts, we implement a mechanism based on a slow-path key derivation (with variable cost key derivation functions, such as PBKDF2 or Argon2), incorporating a deliberate processing delay on the supplicant. Finally, we introduce a ticket-based mechanism to facilitate efficient re-authentication for known devices, bypassing expensive exchanges while preserving system availability under adversarial conditions. Experimental results demonstrate that this architecture significantly mitigates DoS risks without compromising legitimate network access.

Sintering is widely used to manufacture ceramics, but coupled densification and grain growth, material-dependent kinetics, and sparse measurements complicate predictive modeling and process design. We present Sinter-PiNDiff, a retrainable physics-integrated neural differentiable framework for predicting density and grain-size evolution. Two neural networks learn densification and grain-growth coefficients within coupled rate equations, while a smooth saturation factor attenuates densification near theoretical density. The same governing structure, network architecture, and training procedure were fitted independently to published data for MgO, Al-doped ZnO, and CaO-doped ThO2. Tests at held-out temperatures and compositions yielded the lowest mean error in all twelve material-metric comparisons against multilayer perceptron and residual network baselines. For MgO, Al-doped ZnO, and CaO-doped ThO2, respectively, density normalized root-mean-square errors were 14.6%, 10.8%, and 14.4%, and grain-size errors using the same metric were 8.6%, 12.1%, and 19.3%. Removing evolving density from both neural-network inputs increased density and grain-size trajectory errors in all three systems and ten of twelve aggregate errors, supporting density-dependent kinetic feedback. Deep ensembles estimated model disagreement, but empirical coverage showed that the uncertainty bands were not calibrated and did not capture all model-data discrepancies. These results establish Sinter-PiNDiff as a retrainable framework for sparse-data prediction and uncertainty-informed selection of sintering conditions.

We present a polynomial-time reduction from the search variant of the linear code equivalence problem (i.e. the search for a linear isometry between the inputs) to its decisional variant. More precisely, given two linearly equivalent codes $\mathcal C_1,\mathcal C_2 \subseteq \mathbb{F}_q^n$, we show how to recover a linear isometry between them by making a polynomial number of queries to an oracle for decisional linear code equivalence. First, we prove that search-Permutation Code Equivalence (search-PCE – the problem of finding a permutation $\pi\in\mathcal S_n$ mapping $\mathcal C_1$ to $\mathcal C_2$) reduces in polynomial time to PCE (i.e. the problem of deciding if there is a permutation map from $\mathcal C_1$ to $\mathcal C_2$) via at most $n^2$ oracle calls on instances of dimension $k$ and length at most $n^2(n+1)/2$. We then extend this approach to linearly equivalent codes: we recover the permutation part of a linear isometry via at most $n^2$ calls to a Linear Code Equivalence (LCE) oracle on instances of the same size, and we give a deterministic polynomial-time algorithm to recover the diagonal part once this permutation is known. Altogether, this yields a polynomial-time procedure to recover a linear isometry from an oracle for decisional LCE. From a linear-algebraic perspective, our results provide an explicit reconstruction of a monomial equivalence between two matrix representations from oracle access to the corresponding orbit membership problem.

We develop several tools to study the local and global structure of the Teichmüller and moduli spaces for complex nilmanifolds. As a starting point, we prove a structure theorem for holomorphic maps between complex nilmanifolds, which allows for an explicit description of the Teichmüller and moduli spaces. We then introduce two period maps, and give general criteria for the existence of a complex structure on these spaces. To illustrate the theory, we prove a global Torelli theorem for various classes of nilmanifolds, including principal torus bundles over tori and almost abelian nilmanifolds. In particular, we show that there are nilmanifolds of arbitrary dimension and nilpotency index for which the Teichmüller space admits the structure of a complex manifold.

We investigate the long-time behavior of small solutions to the one-dimensional Hartree equation with the soft Coulomb interaction. The Hartree equation arises as an effective mean-field model for the evolution of many-body quantum systems, while the soft Coulomb potential provides a regularized one-dimensional approximation of the Coulomb interaction that removes the singularity at the origin while preserving its long-range character. Owing to the slow spatial decay of the interaction kernel, the equation exhibits long-range nonlinear effects. We prove global existence and modified scattering for sufficiently small, exponentially localized initial data. A notable feature is that, in contrast to the logarithmic phase correction arising in the well-known three-dimensional Coulomb Hartree equation, the one-dimensional soft Coulomb interaction produces a leading phase correction of order $(\log t)^2$, followed by a lower-order logarithmic correction. The proof is carried out in analytic function spaces. The exponential localization of the initial data is converted into analyticity of the transformed profile, and a method based on a generator function with a time-dependent radius of analyticity is used to compensate for the derivative loss arising from the nonlinear phase equation.

Detectors of AI-generated images are typically trained using samples from all Generative AI architectures they must catch, and struggle as soon as a new architecture emerges. Recent approaches have explored self-supervised pre-training as an alternative solution, yet standard frameworks work against the forensic task, e.g., their augmentations overwrite the micro-statistics of image formation. This paper introduces Forensic Twins, a Self-Supervised Residual Learning (SSRL) framework whose pretext task suppresses macroscopic content availability. Each image is mapped through a frozen, off-the-shelf forensic residual extractor, from which two spatially disjoint crops are drawn. Sharing no pixel, the two views retain minimal semantic structure to align, leaving a redundancy-reduction objective with a predominant common signal: the stationary fingerprint of the image acquisition pipeline. Additionally, Forensic Twins is trained exclusively on real images; no AI-generated image is observed at any stage. Experiments show that Forensic Twins attributes AI generator sources with 56.61% accuracy, i.e., 6.13% above the previous state-of-the-art zero-shot method at 375x lower latency. We also demonstrate that fitting a Gaussian Mixture Model (GMM) offline using only the real image embeddings extracted from Forensic Twins turns it into a state-of-the-art zero-shot detector, reaching 97.99% AUC across 27 unseen AI generators, including GANs, diffusion models and commercial systems. Code, weights and exact splits will be made publicly available

The decision problem is whether valence among the users who write reviews of an app has changed enough to warrant investigation of a release, outage, or support issue. We propose a statistical specification for that decision, applied to written Google Play reviews collected under a declared locale, language, sort order, retrieval cap, and window rule. That endpoint returns a selected collection rather than a random sample, so the index describes review writers under the declared protocol and not all app users. What is new is not a sentiment classifier but an auditable interface: an ordinal threshold model is the preferred representation of star ratings, a calibrated two-signal generalized least-squares estimator is retained as an interpretable linear approximation when held-out labels support its conditional-unbiasedness assumption, the estimand is declared before the estimator, and every downstream module–uncertainty, diagnostics, and regularization–is tied to that declaration. The framework separates the latent average of a fixed collected set from the mean of a protocol-defined population of review writers, so measurement and between-review variation are not conflated, and treats helpfulness and recency weights as a different policy target. Shrinkage, cross-app empirical Bayes, histogram and dependence diagnostics, and dynamic smoothing are specified as separate modules with defaults and mandatory sensitivity analyses. A reference implementation, a declared 6000-review three-app collection, and a synthetic stress test exercise the displayed equations: fusion reduced error in clean and small-window scenarios, coordinated contamination caused severe bias and interval failure, and the collected sample reproduces the abstention rules. No real-review validation is claimed.

Numerical inflation targets anchor beliefs. Across euro-area and US professional forecasts, inflation swaps and options, and realized inflation, uncertainty about inflation is compressed at the announced number and kinks exactly there. This paper identifies a cost of the same design that, to our knowledge, has not been shown before, and that appears in second moments only. Within the workhorse New Keynesian model, tolerating part of the inflation a supply shock produces is optimal, yet the optimal tolerated share is not identified: optimal look-through and an unwarranted drift of the effective target are observationally equivalent in the inflation history, so a central bank cannot demonstrate that a warranted deviation is warranted, ex post as much as in real time. Agents holding finite, heterogeneous patience then generate predictive variance that is flat below the target and rises linearly with the expected overshoot above it – an observational-equivalence bill, zero at the announced number and accumulating with the point-years inflation spends above it. The first-order benefit of the number stands; what the cost changes is how inflation uncertainty must be measured. The distance from target explains 70% of the variation in professional forecast variance, and the bill lies within that component; Normalized Uncertainty – to our knowledge the first such correction – removes it. The purge changes inference substantially: on French loan-level data, raw dispersion is unrelated to corporate loan rates, while one standard deviation of the purged measure is associated with rates 89 basis points higher, and cross-country growth and time-series results shift similarly. Conventional measures of inflation uncertainty partly record inflation's distance from its anchor rather than uncertainty about the outlook.

We present Muslim, a production Arabic voice AI platform serving grounded, sourced Islamic knowledge to real users. Beyond a real-time voice pipeline (NeMo Arabic ASR, an OpenAI-compatible LLM endpoint, self-hosted TTS) and a deterministic multi-source retrieval layer routed across six Model Context Protocol servers, we report three things a research prototype typically lacks. First, a released family of fine-tuned Arabic Islamic model artifacts: an efficient tool-routing LLM (Muslim-6B-PRO, 5.94B parameters) and a Modern Standard Arabic TTS model (Fasih-TTS-V1) that ranks 5th of 17 overall and 2nd of 11 open-weight systems on the community-voted Arabic TTS Arena for MSA. Second, an account and metering layer - a free per-account turn allowance, capacity-aware refusal, and email verification deferred to the point it actually matters - that turns an open demo into an operable, abuse-resistant product. Third, a three-layer observability stack (liveness, error reporting, product analytics) built specifically around the system's characteristic failure mode: a GPU-bound agent host going silent while the web tier keeps serving normally. We report real, measured latency and accuracy figures (98.4% recitation-validation accuracy on 124 cases; end-to-end voice latency of 0.9-1.7s) and discuss the concrete engineering trade-offs and limitations of running an Islamic-knowledge voice product in production.

Among the various major subtypes of Type Ia supernovae (SNe), the luminous 1991T-like (Filippenko et al. 1992; Phillips et al. 1992) and 1999aa-like (Krisciunas et al. 2000; Garavini et al. 2004) events stand out as particularly important for constraining the explosion mechanism. Their white dwarf (WD) progenitors are thought to have masses close to or exceeding the Chandrasekhar limit. We present spectrophotometric and polarimetric time-series observations of the overluminous Type Ia SN2019vrq. Between -9 and +12 days relative to the light-curve peak, its continuum polarization is consistent with zero, as seen in most normal-bright SNeIa and despite a possible rise up to 0.2% toward longer wavelengths as seen in subluminous SNeIa. Polarization across major spectral features is marginal ($\lesssim$0.3%) with the exception of the high-velocity CaII near-infrared triplet, which reaches $\sim$0.5%. This low polarization indicates a substantially spherically symmetric electron-scattering photosphere and minimal large-scale ejecta asymmetry. Early-time spectra reveal hot, doubly ionized ejecta dominated by SiIII and blended iron-group features. Intermediate-mass-element lines, such as CaII and SiII, are more clearly seen than in 91T-like events, though these lines remain weaker than in normal SNeIa. The intrinsically higher peak luminosity of overluminous 91T/99aa-like SNeIa (Yang et al. 2022; Phillips et al. 2024), their comparatively short formation timescale for a massive WD progenitor - a channel that becomes more common toward high redshift, and the moderate viewing-angle dependence of their spectrophotometric properties as inferred from their persistently low polarization, together suggest that this subclass may serve as a promising standard candle at high redshift.

We present ClearGS for 3D Gaussian Splatting (3DGS) from handheld videos with uneven viewpoint coverage and mixed frame quality. Rather than selecting frames with binary decisions, ClearGS uses Reliability-aware View Allocation (RVA) to assign graded raw-supervision weights based on appearance reliability, degradation risk, and geometric utility, while weakly reactivating useful suppressed frames to maintain trajectory coverage. Since weighting cannot restore details lost to blur or distortion, ClearGS further introduces Render-Guided In-Video Restoration (RIVR). The current 3DGS render provides a pose-aligned structural candidate, a frozen no-reference restoration expert restores the corresponding raw video observation without any clean reference image, and no-reference perceptual scores select among the render, restored observation, and high-frequency fused candidate. ClearGS then applies Full-Trajectory Repair Consolidation to revisit accepted repairs and preserve details introduced early. On GS2E and GSOTM, ClearGS achieves state-of-the-art overall performance, with consistent CLIP-IQA and MUSIQ gains and LPIPS reductions in most degradation settings, without paired sharp supervision or matched clean references.

This study evaluates a mathematical model of syllable production using an EMA dataset from a recently published study, which proved highly compatible with the model's architecture. The data set consists of regularly structured French phrases of equal duration and reduced phonetic and syllabic complexity. A dedicated procedure was used to transform EMA recordings into Maeda parameters. The Model-generated trajectories were then realigned with these transformed data using Canonical Time Warping (CTW). Statistical validation was conducted via a permutation test, comparing alignment quality against a surrogate condition with phonetically incongruent model outputs. The results show significantly better alignment for phonetically congruent pairings, revealing a strong structural correspondence between the model and the articulatory data. These findings provide experimental support for the representational validity of the model.

Urban uncrewed aerial vehicle (UAV) vision-language navigation (VLN) requires agents to follow instructions across extended urban spaces, inherently demanding long-term memory and geospatial grounding. However, scaling existing benchmarks remains difficult because of their reliance on costly reconstructed 3D assets, limiting geographic diversity and episode scale. To address this, we introduce SatNav, a scalable, long-horizon UAV VLN benchmark built from high-resolution satellite imagery. SatNav targets city-level navigation missions and uses satellite crops as approximations of UAV nadir views for visual observations. Through an automated cue-to-episode pipeline, SatNav constructs 118K episodes from 59 scenes across 18 cities, with an average trajectory length of 379 m. To stress-test long-horizon memory and geospatial reasoning, SatNav defines three task families: Boundary, Landmark, and Route, targeting loop progress tracking, landmark-based spatial grounding, and route following with counting cues. Benchmarking classical VLN agents and recent agents based on large vision-language models (LVLMs) on SatNav shows that city-scale navigation remains challenging. We further introduce SwiftVLN, a modular framework with switchable memory components, and conduct systematic memory-design ablations. Finally, satellite-to-UAV transfer experiments show that satellite-trained navigation models can operate on real-flight UAV observations, showing the practical relevance of SatNav. Our project page: https://eku127.github.io/SatNav/

Large language models (LLMs) exhibit substantial performance disparities between high- and low-resource languages. Beyond lower task performance, they often fail to capture the cultural norms and values of underrepresented communities. In this work, we present the first systematic evaluation of cultural awareness in LLMs for Haitian Creole, a language spoken by millions but severely underrepresented in digital resources. We assess cultural awareness along four complementary dimensions—specificity, bias, diversity, and variation—using a benchmark of culturally salient prompts curated by native speakers in a text infilling setting. Our results reveal a clear gap between cultural awareness in Haitian Creole and higher-resource French, with Haitian performance being more uneven across domains and more affected by French linguistic interference. Story generation further reveals recurring portrayals of Haitian characters through hardship and resilience, showing that even positive characterizations can encode stereotypical narratives. Our code, benchmark, and evaluation framework are publicly available.

Despite the growing capabilities of large language models (LLMs), prompt design remains largely heuristic and ad hoc. This project will explore $\textit{prompt minimization}$, the process of reducing prompts to their smallest, most information-dense form while preserving output fidelity. Practically, shorter prompts reduce computational overhead and inference latency, especially when large contexts, such as entire documents or codebases, are included unnecessarily. Further, longer prompts can damage LLM reasoning and accuracy. Theoretically, the existence of multiple prompts yielding equivalent outputs suggests a high degree of redundancy in the input space, raising fundamental questions about what information is essential to elicit specific model behaviors. We propose three variant frameworks to identify and evaluate minimal prompts and demonstrate that minimal prompts often produce outputs comparable to those of their longer counterparts. These findings suggest new directions for efficient prompt engineering and deepen our understanding of input compression in LLMs.

We study the spectrum of two-dimensional Yukawa theory using Hamiltonian truncation, with a focus on bound states. We extend the framework of Hamiltonian truncation effective theory to include fermions and logarithmic UV divergences. With these corrections, the energy levels converge as a function of the cutoff with the scaling predicted by power counting. We identify a strategy to follow the evolution of fermion-antifermion bound states from weak to strong coupling, and investigate its binding energy, finding good agreement with finite-volume perturbation theory at weak coupling. Finally, we explore the heavy-scalar regime toward the massive Thirring/Sine-Gordon limit, recovering the expected weak-coupling behavior and outlining how the calculation can be extended toward this limit.

Solar analogs are useful laboratories for tests of stellar structure and evolution, particularly in the context of the upcoming PLATO mission. Our study focuses on six seismic solar analogs observed by Kepler and K2. We investigate the impact of the model physical ingredients on these six targets in order to evaluate how well one can satisfy the PLATO precision requirements for stellar parameters with the data quality that is currently available to us. We used as reference values the stellar mass, age, and chemical composition obtained with the Forward and Inverse COmbination (FICO) procedure, based on OPAL opacities and a combination of NACRE II and Solar Fusion II (SFII) nuclear reaction rates. The efficiency of macroscopic transport is calibrated on lithium abundances derived from the analysis of HERMES spectra. From there, we carried out extensive tests of the physical ingredients with the help of a Levenberg-Marquardt local minimisation method using both classical and seismic constraints. In particular, we investigated the impact induced by the use of OPLIB opacities and SFIII nuclear reaction rates on the stellar fundamental properties, as well as the use of various seismic constraints. We quantify the potential differences that changes in the modelling input physics may lead to, and compare their extent with the requirements expected for the PLATO mission (15% in mass, 2% in radius, and 10% in age), providing crucial constraints on the accuracy and robustness with respect to systematics of the determination of fundamental properties for PLATO. For our sample of solar analogs, we are able to accurately retrieve stellar masses within 2.5%. As for stellar ages, we quantify systematics of the order of 10-14% at most, but there are also very promising cases for which deviations remain within 4% depending on the stars and constraints considered.

We demonstrate a simple and efficient method to parallelize and distribute Trotterized Hamiltonian simulation of fermionic systems across multiple QPUs. Using combinatorial covering designs to define a fixed set of fermion-qubit encodings, independent of system size, we demonstrate communication cost scaling as $\mathcal{O}(M q^3 r)$ for a system of $M$ fermionic modes and Trotter number $r$. This improves on the static encoding bound for $q$ QPUs, $\mathcal{O}(M^4 q r)$. In terms of QPU register size $R$, at most $R^{1/3}$ QPUs are able to parallelize at most $R^{4/3}$ fermionic modes. We compare this approach to dynamic encoding using a randomized method, Pauli-weight based optimization and hypergraph partitioning. Applying these to the Hamiltonians of a range of molecular systems split across two and three QPUs, we find the combinatorial covering approach results in significantly reduced communication cost.

The goal of personalized immunogenomics is to recover an individual's germline immunoglobulin gene segments from 'repertoire sequences' altered by trimming, extension, and mutation. In this work, we study the fundamental trace complexity (or the number of samples/traces required for accurate reconstruction) of D-gene reconstruction under three biologically motivated trace-generation models introduced by Bhardwaj et al. (2021) and develop practical algorithms for reconstruction problems left open in the original work. First, for the TrimSuffixAndExtend model, we establish the optimal trace complexity to be {\Theta(n)}, and develop a low-complexity Prefix-Filtered Mode (PFM) decoder that achieves this scaling. Second, for the closely related two-sided TrimAndExtend model, we show that the optimal trace complexity is instead {\Theta(n^2)}, and is achieved by the simple Bit-Wise Mode (BWM) decoder. Third, for the SuffixExtend-t(TrimSuffix) model, we establish polynomially separated lower and upper bounds on trace complexity. Our results follow from information-theoretic lower bounds coupled with tight analyses of the proposed reconstruction algorithms.

Interaction between Rydberg excitons in core-shell Cu$_2$O nanostructures and surface plasmons is investigated. An enhancement of a quadrupole transitions to the Rydberg exciton states in copper oxide plasmonic nanaostructures is examined. A possibility of a modification of plasmonic enhancment (i.e., an average amplification factor) of electric field distributions in the vicinity of nanostructures, which depends on its shape and geometrical size, is also discussed.

In Mueller's dipole cascade without transverse dimensions, with one counted particle for each dipole produced in a window, the multiplicities in three consecutive rapidity windows are Poisson counts of one Gamma-distributed source. We show that at fixed middle multiplicity the outer two share less information than the Gaussian value $-\tfrac12\ln(1-\rho^2)$ of their partial correlation, a value that second moments alone determine. For windows of one common width at a constant splitting rate this value never exceeds $\tfrac12\ln(4/3)$. The Gaussian value is not a bound on mutual information in general, and for a source of the same mean and variance with another law the information can exceed it. The law of the counts keeps its form under migration of particles across window edges and detection losses when both act on each particle independently of the others and of the source, and its factorial cumulants of second and third order test necessary conditions for it. In Monte Carlo simulations of the cascade with transverse dimensions at leading logarithmic accuracy and fixed coupling, the counts are not Poisson counts of one source, so the bound is a result of the model without them. In the model without transverse dimensions the law also fixes the normalized second factorial cumulant of one window at $1/k$, with $k$ the number of initial dipoles. The values of this cumulant formed from the mean and variance that H1 publishes for deep inelastic scattering lie far below the value one of a cascade from a single initial dipole with one counted particle for each dipole produced in a window. The quantity bounded is a classical conditional mutual information between counted multiplicities.

Search is one of the most important features in e-commerce, directly driving customer engagement and business growth. A good product search system must show both relevant and desirable results. However, retail search presents unique challenges. User intent can range from exact matches to open-ended discovery. Search systems must also balance multiple goals, such as relevance, revenue, and profit, while keeping response times low. Traditional keyword-based methods often fall short in handling natural language or semantic queries. Vector search helps alleviate these issues, but it can miss key intent signals or return low-precision results. In this paper, we present the design of a hybrid search system at Target that combines lexical and vector search. We describe our approach to data processing, embedding training, precision control for the final result set, multi-channel result fusion (where we compared fusion strategies and adopted weighted interleaving), and the performance optimizations used to maintain low latency for production deployment. Our method improves offline evaluation metrics, and in online A/B testing it raised click-through rate by 0.97%, order conversion by 0.98%, and demand per visitor by 1.10% over lexical-only search, while roughly halving zero-result searches. The resulting system is deployed at scale and serves millions of guests daily.

Use of the Kohn-Sham eigenvalue gap (highest occupied molecular orbital to lowest unoccupied, HOMO-LUMO) as a fundamental gap (I-A, I= ionization potential, A= electron affinity) estimator is a long-standing challenge for relatively simple exchange-correlation functionals, e.g. meta-GGAs. The LAK meta-GGA [T. Lebeda et al. Phys. Rev. Lett. 133, 136402 (2024)] was developed to emphasize certain constraints in ways intended to make the LAK KS gaps close to the HSE06 hybrid functional [A.V. Krudau et al. J.Chem. Phys. 125, 224106 (2006)] gaps. The LAK functional was tested against a small atomization data set and a very small bond length set, with no molecular HOMO-LUMO gap testing. Here we test LAK against widely used, large reference data sets for molecular heats of formation, bond lengths, harmonic frequencies, I, and A and compare with r2SCAN [J.W. Furness et al., J. Phys. Chem. Lett. 11, 8208 (2020)] as well as HSE06 and PBE0, a quite different hybrid. On these tests LAK does not give meaningfully better geometries, heats of formation, HOMO-LUMO gaps, or DFT IP theorem satisfaction than r2SCAN. Compared against experimental I and A values the same holds; LAK results resemble r2SCAN results, not HSE06. Moreover, the HOMO and LUMO shifts generated by the NCAPR GGA [J. Carmona-Espindola et al., J. Chem. Phys. 157, 114109 (2022)] as derivative discontinuity estimates, correct the bare NCAPR HOMO and LUMO eigenvalues to substantially better agreement with experiment for both fundamental gap and IP theorem than do the meta-GGAs (LAK, r2SCAN) or hybrids (HSE, PBE0). We discuss possible reasons why LAK behaves differently for molecules and solids.

The influence of crystallographic orientation on the early stages of the oxidation of pure Zr single crystals was studied by X ray photoelectron spectroscopy (XPS). Two samples, cut from the same a-Zr single crystal, were oxidized simultaneously at room temperature and O2 pressures of 1 X 10-8, 1 X 10-7 and 1 X 10-6 torr: one with its surface normal parallel to the c-axis (Z1, basal (0001) plane) and the other with its normal 12deg from that of a prismatic plane (Z2, near-prismatic orientation). At all three pressures the oxidation kinetics of both samples followed a three stage, logarithmic type behaviour, but the near prismatic sample (Z2) incorporated oxygen faster and to a greater extent than the basal sample (Z1) in every case. Deconvolution of the Zr 3d core level spectra resolved, in addition to metallic Zr and ZrO2, two sub-stoichiometric Zr O compounds, denoted (ZrO)a and (ZrO)b, with binding-energy shifts of 1.3 and 2.3 eV with respect to metallic Zr, consistent with sub oxides previously reported for polycrystalline Zr. Angle resolved XPS showed that ZrO2 is the outermost compound in both orientations, while the two sub-oxides are distributed nearly homogeneously through the film. Oxide thicknesses, calculated from the attenuation of the metallic Zr 3d signal, ranged from 10 to 13 A for Z1 and from 14 to 19 A for Z2 depending on the oxidation pressure, in close agreement with earlier measurements on cold worked polycrystalline Zr, whose surface texture is dominated by prismatic-oriented grains.

Cosmological first-order phase transitions are well-known sources of gravitational waves from bubble collisions and plasma dynamics. In this work, we investigate an additional microscopic source directly arising from the associated production of particles and gravitons during bubble collisions. Focusing on runaway bubble walls, we calculate the gravitational wave spectra from two-body scalar-graviton production and three-body processes involving scalar or fermion pairs accompanied by a graviton. We derive analytic approximations and numerically evaluate the resulting spectra, revealing a high-frequency gravitational wave component that can complement conventional lower-frequency signals from the same phase transition. This mechanism offers a potential multiband signature and establishes a connection between gravitational radiation and the production of dark matter or other new particles. Our findings provide a new physics target for future high-frequency gravitational wave experiments and a complementary probe of particle production in the early Universe.

Federated Learning (FL) allows devices with private data to collaborate in training a shared model. We present a next-generation FL system based on Trusted Execution Environments (TEEs) that addresses operational challenges associated with earlier systems and provides externally verifiable central Differential Privacy (DP) guarantees for the first time while offering a better privacy-utility tradeoff. In our system, devices upload data encrypted with keys managed by a TEE-hosted Key Management Service (KMS). The uploaded data is cryptographically tied to a policy limiting the set of Python programs that may later process the data in server-side TEEs. External parties may inspect public transparency logs to observe the set of workloads allowed by these policies. Our experimental results show that the new system improves device coverage and favorably shifts privacy-utility curves by enabling collected data to be integrated into the server-side workload at a schedule that optimizes DP guarantees and is unaffected by device availability. Our new system has been productionized, enabling models for the Android Keyboard (Gboard) to be trained faster and achieve better accuracy under smaller, now externally verifiable privacy budgets in comparison to models trained using the prior system.

We establish sharp weak-type bounds for commutators of Calderón–Zygmund operators and $\text{BMO}(\mathbb{R}^n)$ symbols with respect to $A_p$ weights. We prove that the bound \looseness=-1 $$ \|[b,T]\|_{L^p(w)\rightarrow L^{p,\infty}(w)}\lesssim [w]_{A_p}^{\max(2,p')} $$ for $p \in (1,\infty)$ and $w \in A_p$ is sharp for general $T \in \text{CZO}(\mathbb{R}^n)$ and $b \in \text{BMO}(\mathbb{R}^n)$. In the case of the standard logarithmic symbol, we show that the estimate improves to $$ \|[\log|\cdot|,T]\|_{L^p(w)\rightarrow L^{p,\infty}(w)}\lesssim [w]_{A_p}^{\max\big(2,\frac{1}{p}+\frac{p'}{p}\big)} $$ and that this dependence is optimal for this symbol. Our upper bound for the logarithmic symbol follows from the pointwise control of the commutator by a sum of the Hardy–Littlewood maximal operator and the squares of the radial Hardy operator and its adjoint, while our lower bounds are consequences of a general result for singular integral operators.

In daily life, people hear speech, footsteps, and music around them. We can often recognize these sounds and judge where they come from. Each sound source can be shown on a separate acoustic map, a rectangular image covering $360^{\circ}$ horizontally and $180^{\circ}$ vertically. The map shows the directions occupied by the source as a region and the sound energy within that region. A class label identifies the sound. Predicting these labeled acoustic maps from audio is called semantic acoustic imaging. Such maps could help robots perceive their surroundings and allow augmented reality displays to show sound regions and classes over the real world. Existing models can recognize sound classes and estimate a direction for each source. However, a direction alone does not describe the source region or its energy. Acoustic imaging must also distinguish sound sources in nearby directions, while the number of active sources and the regions they occupy can change over time. We therefore propose the Semantic Acoustic Imaging Detector (SAID), which predicts a separate labeled acoustic map for each active source from audio. First, we pretrain Audio2Sph, SAID's audio encoder, through sound energy estimation across directions without class labels. Then, we train the complete SAID model to predict source regions, energy, and classes together. We also develop a pipeline that generates simulated recordings for pretraining and supports fine-tuning on real recordings. On the official DCASE2026 Task 3 Track A evaluation set, our submitted system ranks first with 0.1080 macro-averaged mean average precision (Macro mAP) and 0.3962 Macro Pearson $r$. Demos and code are provided at https://github.com/IN03X/SAID.

Forecasting short-horizon mid-price movements from limit order book (LOB) data is central to algorithmic trading, yet most deep LOB forecasters are point predictors: they output a direction or a displacement, but never indicate which of their forecasts can be trusted. We introduce UQ-LOB, a lightweight, encoder-agnostic uncertainty quantification module that attaches to any pretrained LOB encoder and, in the spirit of attentive neural processes, conditions each forecast on a context set of recently completed windows whose outcomes are already realised. The UQ-regression variant outputs a calibrated Gaussian over the future tick displacement, while the UQ-classification variant outputs a categorical distribution over down/up/stationary. Both expose a scalar confidence (predicted signal-to-noise ratio or class probability) that supports selective prediction. On 5.2 billion LOB events across seven cryptocurrency assets and horizons of 5, 10 and 15 seconds, UQ-regression attains near-nominal 68% interval coverage, and restricting to the most confident 10% of predictions raises directional macro F1 by 0.11-0.15 for UQ-regression and 0.05-0.11 for UQ-classification, at every horizon. On large, economically meaningful moves, the tightest confidence tier reaches a directional F1 of 0.88 (down) and 0.83 (up) at the 5-second horizon.