Papers

The estimation and analysis of facial expression intensity play a crucial role in affective communication and human-computer interaction. Previous research has primarily focused on detecting and estimating facial expression intensity from frame-level 2D representations. However, this limitation restricts a comprehensive understanding of real-world facial expressions, as they are inherently 3D and temporally continuous. This paper investigates the perception of facial expression intensity by introducing the 4D Facial Expression Intensity Dataset (4DFEID). We employ a parametric face model and compile a total of 2,869 mesh sequences with controlled geometric variations, generating 4D data instances with diverse peak intensities and identity attributes. Using a Likert scale, we collect more than 90,000 subjective intensity perception ratings via a crowdsourcing platform. We explore various architectures and aggregation methods to establish baselines for episode intensity estimation on the new dataset, revealing that spatial-temporal graph models consistently outperform traditional frame-aggregation methods. In contrast to existing datasets that rely on 2D static imagery, the proposed 4D-FEID dataset provides the community with a unique and vital resource for investigating the perception of facial expression intensity through the use of dynamic 3D stimuli. By offering high-fidelity, spatio-temporally coherent facial data, 4D-FEID establishes a new foundation for research into more nuanced and naturalistic expression analysis, thereby addressing a gap in the current landscape of affective computing and human-computer interaction studies. The dataset is available at link.

The low-lying $1/2^+$ state of $^9$Be has been interpreted either as a near-threshold resonance or as an $^8$Be+$n$ virtual state. Because both pictures can produce a pronounced threshold enhancement, its nature is difficult to determine from energy-domain observables alone. We investigate the low-lying $1/2^+$ state through its real-time decay dynamics and compare its behavior with that of the well-established $5/2^+$ resonance. A wave packet produced by electric-dipole excitation of the $^9$Be ground state is propagated in an $\alpha+\alpha+n$ three-body model. We analyze the survival probability, the fit-window dependence of the effective decay width, and time-dependent densities in different Jacobi-coordinate representations. The $1/2^+$ survival dynamics are further compared with an $s$-wave effective-range description motivated by the dominant $^8$Be+$n$ component found in our previous three-body studies. The $5/2^+$ channel shows approximately exponential late-time decay, with an effective width of $0.264$ MeV, close to the complex-scaled pole width of $0.285$ MeV. In contrast, the $1/2^+$ effective width decreases as the fitting interval is shifted to later times, and no stationary decay width is obtained. Its late-time survival probability is well described by the effective-range spectral distribution. The density evolution develops an $^8$Be+$n$-like asymptotic configuration while retaining a localized $p_{3/2}$ $\alpha n$ component suggestive of a $^5$He+$\alpha$-like configuration in the internal region. The $1/2^+$ decay differs qualitatively from resonance-like exponential decay and is consistent with near-threshold virtual-state dynamics. The distinct internal and asymptotic configurations revealed by the time-dependent densities provide complementary structural information beyond that contained in a single decay width.

Zero-determinant (ZD) strategies are a class of strategies in repeated games, which unilaterally control payoffs. It has been shown that several ZD strategies promote cooperation in social dilemma games. It has been widely believed that the performance of ZD strategies is expressed as “unilaterally enforce linear relations between payoffs”. However, in order to interpret properties of ZD strategies in repeated games without discounting, previous studies assumed that the limit of the Cesaro average of the probability distribution of the action profile exists. Here, we explain the performance of ZD strategies without this assumption, which leads to a stronger result than previous ones.

We show that adaptive Bayesian estimation in fluorescence state detection reduces to a comparison of elapsed time with two arithmetic time sequences fixed by the bright and dark count rates and confidence parameters $\epsilon_d$ and $\epsilon_b$. No probabilities are computed in real time and the resulting implementation, which we term sequential deadline detection, is provably optimal in mean detection time. The analysis permits closed-form expressions for detection-time distributions, mean detection times, error rates of both channels accounting for miscalibrated rates, decay between the states during detection, and afterpulsing in single photon counting modules. The predictions are validated by Monte Carlo simulation and demonstrated by fluorescence detection with Ba$^+$

We identify a common action–density structure for the Hartree and Vlasov–Riesz equations and their coupling. Along free rays, the feedback between nonlinear actions and rescaled densities governs wave phase modulations and kinetic momentum translations. This structure persists under the exchange of sources, providing convergent modified profiles and an explicit recursive construction of the asymptotic corrections. We thereby establish small-data global existence and modified scattering in three spatial dimensions for the Hartree equation, the Vlasov–Riesz equation, and the coupled Vlasov–Hartree system, with interaction potential $\lambda|x|^{-\alpha}$, $\lambda\in\mathbb R$ and $0<\alpha\leq 1/2$. In this strongly long-range regime, the asymptotic action contains finitely many successive corrections beyond its leading term. To our knowledge, these are the first forward modified-scattering results for all three models throughout this range.

We study Scott-product compatibility and sobriety for countable meet-continuous directed-complete partial orders (dcpos). We construct a countable meet-continuous dcpo $\mathcal R$ whose Scott space is well-filtered but not sober, although the Scott topology on every finite power is the product of the factor Scott topologies. The ideals of $\mathcal R$ admit a unique finite-parameter description. We also construct a countable meet-continuous dcpo $\mathbb{H}$ with a greatest element whose Scott space is coherent and well-filtered, but whose Scott square has a topology strictly finer than the ordinary product topology. These constructions answer negatively two questions about countable meet-continuous dcpos. On the positive side, we prove that a meet-continuous $L$-dcpo, meaning a dcpo whose principal ideals are complete lattices, is Scott sober whenever its Scott square carries the ordinary product topology. This answers Jia's core-compactness question affirmatively for $L$-dcpos without any countability assumption. Finally, we characterize well-filteredness of $L$-dcpos by boundedness of closed Rudin sets and formulate related open questions.

Removing an input-scaling module changes both a classifier and the perturbations reaching its encoder. A robustness difference can therefore reflect the comparison rule as well as the module. We demonstrate this problem in a four-qubit quantum-attention detector on generated power-grid trajectories. A learned scaling module appears beneficial at a fixed physical attack budget, but matching an upper bound on perturbations at the encoder reverses the ordering. Neither comparison alone establishes a robustness benefit caused by the module. The initial test also perturbs clean examples into attacked examples while retaining their original labels; tests restricted to already attacked examples do not establish a benefit. Replacing a trained model's input scales disrupts detection. Retraining its linear classification layer restores the detection rate, but changes individual predictions, leaving the comparison descriptive rather than causal. Two further design checks explain why the input quantum Fisher information regularizer cannot train this model's query parameters, and why removing confidence bounds does not establish a larger certified radius. The evidence is limited to ten seeds, exact simulation, synthetic data, and a restricted set of attacks; classical baselines achieve better clean prediction. The practical lesson is to specify which perturbation budget is fixed, check that attacks preserve labels and interventions preserve predictions, and distinguish exploratory controls from confirmatory evidence.

The rapid deployment of broadband low Earth orbit (LEO) constellations provides new opportunities for next-generation navigation. Compared with conventional global navigation satellite systems (GNSS), broadband LEO systems provide stronger received signals, wider bandwidths, rapidly varying satellite geometry, and larger constellation scales, offering new capabilities for positioning, navigation, and timing (PNT). Meanwhile, their communication-oriented waveforms, high dynamics, and resource-constrained architectures introduce new challenges that are fundamentally different from those in GNSS. This article provides a comprehensive overview of broadband LEO constellations for next-generation navigation, covering their fundamental characteristics, representative use cases, key enabling technologies, and open challenges. We first compare broadband LEO systems with conventional GNSS and summarize the main technical routes for LEO-based PNT. We then discussed the emerging use cases in GNSS-challenged, multipath-dominant, and integrated communication and navigation (ICAN) scenarios. Key enabling technologies including waveform design, high-dynamic signal processing, and constellation resource scheduling are further reviewed. As a representative case study, the navigation-oriented optimization of the Starlink primary synchronization sequence demonstrates that subcarrier power allocation can improve multipath resistance while maintaining ranging accuracy. Finally, several challenges and future directions of broadband LEO-based navigation are discussed, aiming to provide reference for future researches.

Urban policies are often evaluated assuming their effects decay with geographic distance. However, cities are systems of human movement, and policy effects propagate through individual mobility rather than spatial proximity alone. This study demonstrates that neglecting mobility can systematically distort estimates of how urban policies affect activity. Using large-scale mobility data, we estimate the causal effects of three transportation interventions in Boston and New York City using both conventional geographic and mobility-based definitions of exposure. Geographic designs detect no effects, or even negative ones. When exposure is defined through mobility networks, the same interventions reveal substantial behavioral change. Boston's free-fare program increases overall activity by approximately 16%, with more pronounced effects among lower-income communities. Open Streets programs significantly increase visits to nearby businesses, particularly food establishments. For congestion pricing, geographic approaches suggest a decline in activity in Manhattan, while a mobility-based perspective indicates no net reduction but a shift away from shopping and grocery destinations. Analysis across 138 metropolitan areas in eight countries further reveals that geographic and mobility-network distances diverge locally in every city, indicating that this discrepancy is a general characteristic of urban mobility. Collectively, these findings show that urban policies operate not only where they are implemented but where and how people move, so their impact must be measured through mobility rather than through space.

Full-duplex voice agents make many small, closed decisions, which current systems answer by slow autoregressive decoding. We propose DuplexJev, which feeds ASR-encoder hidden states through a small connector into a frozen LLM and reads each question as a single-token distribution over its options. Nothing is decoded, and an 8-GPU node answers 80 decisions about eight utterances in about 0.1 s. With a last-layer connector, spoken QA stays close to reading the transcript (90% vs. 91%). DuplexJev also hears the speaker: gender and emotion accuracy both reach 90% (from 55% and 28%) with a cross-attention connector, whose spoken QA drops by only 1 point (83% to 82%). We train decisions with cross-entropy on the read-out answer token, instead of the usual transcript distillation, whose teacher never hears the voice, and keep distillation for content. Encoders and LLMs are interchangeable; we release weights, training recipe, a batched-inference pipeline for full-duplex serving and a bilingual spoken-QA set.

The ability to derive the governing equations of a dynamical system from data is essential for prediction and control across many scientific fields. Two challenges, however, are the expansion of candidate terms when reducing higher-order equations to first-order form and noise from estimating derivatives. We propose Higher-Order-KANDy, which addresses both problems by using delay and derivative ("jet") embeddings of the original data with derivative-ordered Kolmogorov-Arnold Networks for Dynamics (KANDy) layers stacked into a machine learning architecture. Compared with SINDy and WSINDy on noisy data, Higher-Order-KANDy learns governing equations without a candidate library (e.g., recovering the full-period $-\sin(u)$ with nothing named in advance), natively for higher-order systems, and holds the correct term structure as noise grows, selecting the exact active set on every seed at noise levels up to 7.5% where an oracle weak-form regression admits a spurious term, though the oracle remains the more accurate coefficient estimator. We leverage the relationship between delay and differential embeddings, along with the Kolmogorov-Arnold representation theorem, to construct a delay-to-jet map and learn it as a Kolmogorov-Arnold network to preserve derivative structure and provide meaningful terms that appear in the governing equation. In our synthetic benchmarks, this construction recovers equations in the presence of coarse sampling and noise.

Quantum Fourier sampling may help audit the spectral learnability of delay-based physical unclonable functions (PUFs). We ask whether that promise survives access matching, a strong classical comparator, and oracle synthesis. Three gates structure the evaluation. Structure: low degree is not small support at reachable challenge lengths; for 4-XOR at $n=14$, degree $\le d_f(0.1)$ admits $91\%$ of all $2^n$ characters and the median $90\%$-mass set spans a third of the spectrum. Algorithmics: constructing the phase oracle logically implies classical membership access, making Kushilevitz–Mansour the correct baseline; across 45 tasks it exhausts each finite domain, and no 4-XOR ideal-sampling case reaches $90\%$ mass within $2^n$ calls. A quantum-kernel diagnostic appears more favorable, with geometric difference rising to $2.151$ at $N=512$ challenges, but it correlates $0.991$ with $1/\sqrt{\lambda_{\min}(K_C)}$ for the classical Gram matrix $K_C$, and the 4-XOR label-complexity ratio does not exceed a balance-preserving permutation null ($p=0.930$). Trace-normalized geometric difference can therefore grow through classical ill-conditioning alone, without task-label alignment. Implementation: a simulator-validated fixed-point phase oracle based on the quantum Fourier transform admits an $18.9\%$ routed-depth reduction, yet the least certified precisions have estimated durations of $1.18$–$1.55\times$ the median dephasing time $T_2$ of the mapped qubits on a static backend snapshot, without hardware execution. We find no end-to-end advantage in the evaluated regime, although ideal sampling does use fewer coherent calls on the thresholded task. The contribution is the Three-Gate Quantum Audit Protocol: a reproducible procedure separating an ideal query advantage from a realizable security benefit. This is not a claim about deployed silicon and not an impossibility result.

In conventional Landau theory, simultaneous continuous onset of superconductivity and the breaking of additional symmetries generally requires a multidimensional order parameter or fine tuning. We show that symmetry fractionalization provides a different route. In a Kondo lattice with a $\mathbb Z_2$ spin liquid, the Kondo fields inherit the projective symmetry action of the spinons. When the projective time-reversal action on these fields squares to $-1$, the projective symmetry algebra forbids a condensate that preserves time-reversal symmetry and enforces twofold degeneracy of the critical modes. Because Kondo condensation in a $\mathbb Z_2$ spin liquid also breaks electromagnetic $U(1)$ symmetry, superconductivity and time-reversal symmetry breaking can emerge at a single continuous transition, even if crystal symmetry admits only one-dimensional irreducible representations. We demonstrate this mechanism in an anisotropic Kondo-Kitaev model, where superconductivity and the breaking of time-reversal and mirror symmetries emerge simultaneously.

Magnetic resonance elastography (MRE) is a noninvasive imaging modality for quantifying the viscoelastic properties of soft tissues from shear wave propagation. Recovering the complex-valued shear modulus from measured displacement fields leads to a severely ill-posed inverse problem, particularly in the presence of noise and limited boundary excitations. We investigate adjoint-based optimization, convolutional neural network (CNN) reconstruction, and a hybrid framework combining both approaches. The forward model is based on a scalar form of the modified stationary Stokes system with a complex shear modulus. We establish well-posedness of the forward problem, existence of minimizers, and first-order optimality conditions for the adjoint-based formulation, and implement a nonlinear conjugate-gradient method with Armijo line search. While PDE-constrained optimization can accurately refine coefficient reconstructions, its performance depends strongly on initialization. We therefore construct a two-dimensional CNN that maps complex-valued displacement measurements to spatially varying complex shear modulus fields and provides rapid, informative initial reconstructions. The proposed hybrid method uses the CNN reconstruction to initialize the adjoint-based optimization, yielding faster convergence and improved accuracy. The CNN is trained on coefficient fields containing individual perturbations and tested on both individual and previously unseen combined configurations. Numerical experiments demonstrate that the CNN generalizes to these more challenging configurations, while subsequent PDE-constrained optimization further refines the reconstructed coefficient. These results demonstrate the potential of combining data-driven initialization with physics-based optimization for efficient and accurate MRE reconstruction.

Covariance estimation is a key component of many applications in system identification and data-driven control. Although heavy-tailed distributions may lack a covariance matrix to estimate, the shape matrix provides a well-defined, scale-free generalization for the broad family of elliptical distributions. In this setting, practitioners often employ Tyler's M-estimator (TME), which is defined implicitly and is computed using a fixed-point iteration. The spatial sign covariance matrix estimator (SSCM) offers a much simpler alternative: it corresponds to one iteration of TME. Yet, the SSCM has been largely regarded as an inferior estimator due to its statistical inconsistency under fixed-dimensional asymptotics. By contrast, using second-order tools from random matrix theory, we establish that, under standard assumptions, SSCM asymptotically dominates TME in Frobenius risk when dimension and sample size grow proportionally. This advantage arises from implicit regularization, which reduces variance at the cost of a bias that vanishes as the dimension grows. Numerical experiments support the theoretical predictions even at moderate dimensions.

As large language models are increasingly deployed for multi-step reasoning, verifying the correctness of their outputs has become essential for maintaining reliability at scale. Verifying the correctness of large language model outputs is often done by querying a costly ground-truth oracle, which is impractical to invoke at every step in an online setting. Prior work addresses this by querying a single weak verifier on every step, and using its score to decide whether the costly strong verifier needs to be queried as well, reserving strong verification for only a small fraction of the steps. However, a single fixed weak verifier may not perform consistently well as the subject matter or difficulty of incoming queries changes over time, and committing to one in advance risks either overly costly or inaccurate verification. We introduce OMVV (Online Multi-Verifier Verification), an algorithm that maintains a pool of $K$ candidate weak verifiers with differing cost and verification performance, and adaptively routes each round's decision to a verifier selected via an online score combiner and an exponential-weights routing policy. OMVV provides a distribution-free, finite-time guarantee on false-accept and false-reject rates across the full pool of verifiers, and further achieves sublinear regret against the best fixed verifier in hindsight under a combined cost and consistency objective. Experiments on reasoning dataset benchmarks show that OMVV achieves higher accuracy at lower verification cost than any single fixed verifier, across a range of operating budgets.

Post-deployment feedback from users can be a cost-effective, scalable, and representative means to monitor and improve generative AI systems and features. When implemented effectively, giving such feedback can increase users' engagement with and trust in GenAI systems. Government regulations and industry guidelines call for post-deployment user engagement, but there is little guidance on designing mechanisms that are usable for consumers and provide actionable input for product teams. We conducted a multi-phase study as a collaboration between academic researchers and eBay. Our benchmark evaluation of current industry approaches identified common issues including lack of discoverability, unclear terminology, and inattention to user value. Based on these findings, we developed best-practice recommendations and designed and tested a prototype feedback-collection tool. The tool aimed to provide users with an efficient, flexible, and positive feedback-giving experience, and provide product teams with rich data on performance and potential problems in a usable format.

In this paper, we investigate how spatial heterogeneity affects the stability of positive equilibria in one-dimensional predator–prey reaction–diffusion systems with Neumann boundary conditions. Our examples show that spatial heterogeneity can lead to instability even when the positive equilibrium is unique. For the smooth, spatially heterogeneous coefficient families constructed here, this equilibrium loses linear stability as a parameter in the reaction coefficients varies. At the critical parameter value, a simple conjugate pair of eigenvalues crosses the imaginary axis transversely, and a Hopf bifurcation produces a local branch of strictly positive, nonstationary classical time-periodic solutions. The examples include two families with spatially concentrated prey, one with an everywhere positive and the other with an everywhere negative predator intrinsic growth rate. A third family has a spatially concentrated predator population and an everywhere negative predator intrinsic growth rate.

AIM. This study examines the PLATO mission sensitivity to small transiting planets (radii between 0.5 and 2.0 R$_\oplus$) using Autoregressive Planet Search (ARPS) methodology based on simulated lightcurves. Methods. Several thousand simulations of PLATO-like lightcurves from magnetically active stars expected from the PLATO Long Observation Phase South 2 (LOPS2) field with injected transiting planets smaller than 2.0 R$_\oplus$ radius are examined. The lightcurves are analyzed with ARPS that uses autoregressive modeling for filtering of nonplanetary signals and the Transit Comb Filter periodogram for planetary transit detection. False alarms of nonexistent periodic signals are reduced by thresholds on two ARPS parameters, peak periodogram power, and transit duration. Results. When applied to PLATO's simulated first three months observations of bright stars (P1 and P2 samples), at least 50% of Earth-size planets are recovered for periods shorter than 40 d. In two years of LOPS2 observation, 25% are recovered for periods less than 240 d. Methodological implications on detrending, periodograms, False Alarms, vetting, and catalog evaluation are discussed. The utility of a positive predicted value heat map after detection and vetting is demonstrated. Conclusions: While the ARPS planet detection procedure is not entirely well-suited for PLATO analysis, this study provides methodological insight into the sensitivity and reliability of small transiting planet detection.

We study the normal-mode stability of a flat interface between compressible Euler flow in the air and incompressible Euler flow in the water, including gravity and surface tension. The base state is quiescent water beneath a hydrostatic, constant-entropy shear flow $U(z)$ with $U(0)=0$, so the atmosphere is neutrally stratified. For a small ratio $\varepsilon$ of the air density at the interface to the water density, and for uniformly subsonic wave-relative flow, we establish a compressible analog of the critical-layer instability of Miles, extending the rigorous incompressible theory of Bühler, Shatah, Walsh, and Zeng. For a real phase speed $c_r$, the singular coefficient at a regular critical level $U(z_c)=c_r$ has a residue proportional to $Q=U''+gU'/a^2=\bar\rho_0(U'/\bar\rho_0)'$, with gravitational acceleration $g$, sound speed $a$, and normalized base-state air density $\bar\rho_0$. Thus the gradient of specific vorticity replaces the curvature $U''$ of the incompressible theory. We prove a limiting absorption principle at regular, non-resonant phase speeds and a residue formula for the imaginary part of the boundary value $Z_{\mathrm a}^+(k,c_r)$ of the air-side pressure response at wavenumber $k>0$. Let $c_k>0$ be the phase speed of the free gravity-capillary wave. If $\mathrm{Im}\,Z_{\mathrm a}^+(k,c_k)>0$, then, for sufficiently small $\varepsilon$, the dispersion relation has exactly one unstable root $c_\varepsilon$ in a neighborhood of $c_k$ that does not depend on $\varepsilon$. A sufficient condition is $Q\le0$ at every critical level and $Q<0$ at the highest one. As the Mach number $M\to0$, $Z_{\mathrm a}^+$ converges to its incompressible counterpart at rate $O(M^2)$. Under the corresponding incompressible sign condition, the unstable root satisfies $|c_\varepsilon(M)-c_\varepsilon(0)|\le C\varepsilon M^2$ with $C$ independent of $\varepsilon$ and $M$.

An email assistant should not complete less work simply because a user phrases the same request differently. Yet most benchmarks test each task with only one canonical request, leaving this form of robustness largely unmeasured. We test whether email assistants remain reliable when the requested information, available evidence, and expected outcome stay fixed, but the communication style or English variety changes. We construct validated variants along five communication-style axes and four rule-based dialect conditions, and evaluate them on three benchmarks: a retrieval-augmented generation (RAG) pipeline and two tool-using agents. Indirect requests reduce performance on all three benchmarks, while formal requests reduce performance on both agentic benchmarks. Examining the systems more closely shows that these failures have different causes. Verbose requests mainly hurt a lexical retriever by making the relevant email harder to find. By contrast, indirect and dialect variants remain harmful even when the relevant email is retrieved. In the agentic setting, indirect and formal requests mainly cause the agents to omit required actions, not to take more unsupported actions. These results show that a successful response is not enough to establish robustness: evaluations should vary how requests are expressed and separately measure whether agents complete the requested work.

Vision-language-action (VLA) models increasingly incorporate intermediate reasoning to improve robotic manipulation, yet existing approaches primarily reason about observed states without explicitly anticipating future scene evolution. Extending such reasoning to explicit future rollouts at every inference step, however, introduces substantial computational overhead. We propose IG-VLA, a VLA reasoning framework that enables models to imagine the future and internalize the gist. Our Latent Spatiotemporal Reasoning learns to imagine task-relevant future scene evolution directly in visual representation space, guiding action prediction without costly pixel-level video generation. To further reduce inference overhead, we introduce Scene Gist Memory, which internalizes reasoning-derived scene-behavior associations into a compact Scene Gist Token, preserving the benefits of future reasoning while bypassing explicit future imagination at inference. Extensive experiments on LIBERO, LIBERO-Plus, and VLABench demonstrate the effectiveness and efficiency of IG-VLA. On the LIBERO-Plus Language suite, both the reasoning and gist policies outperform the strongest baseline by nearly 6% in success rate. The gist policy also achieves up to 6.38x speedup over baselines, reducing inference latency from 1081ms to 169.5ms per action chunk on a single NVIDIA A6000 GPU. These results demonstrate that future spatiotemporal reasoning can be effectively internalized for efficient VLA deployment.

Quantum Fisher information adds over independent probes, so a single parameter is best measured probe by probe. We show that Gaussian measurements, the linear optics and homodyne detection of optical and microwave experiments, break this rule: two independent modes are better measured together. The uncertainty principle leaves a linear detector half of phase space, and for two modes the choice of half is a resource. Bell homodyne, a balanced beam splitter followed by two homodyne detectors, beats the best separate readout when the modes differ in quadrature width and in how the width responds to the parameter. Heterodyne detection splits a mode on a beam splitter to read both quadratures and pays one unit of vacuum noise for the empty port. Bell homodyne fills that port with the second mode, so the noise turns into signal. We prove that the gain stays below $(\sqrt2-1)^2=17.157\%$ for every parameter carried by widths. For thermal modes at one temperature, the canonical case, a gain was conjectured impossible. We prove that any frequency difference opens a temperature window and that the gain peaks at $12.699\%$ at frequency ratio $3.318$, where Bell homodyne is the optimal Gaussian measurement. Standard hardware reaches the gain: two coupled resonators with two homodyne detectors, or a phase-preserving amplifier whose idler band is fed by the same thermal source.

Intermittent computing enables batteryless embedded devices to operate in harsh environment, but frequent power failures interrupt program execution and require careful management of state across power cycles. The core challenge is to guarantee both memory consistency and forward progress while minimizing the energy overhead of checkpointing. Recent compile-time approaches identify regions that fit in the energy buffer. At runtime, the device waits and recharges between the regions. However, they still rely on greedy or path-local heuristics that commit to local decisions and can miss globally lower-overhead boundary placements. To address this limitation, we present Bao, a system that jointly finds optimal energy-aware region formation and memory allocation decisions with formal correctness guarantees, by formulating it as a mixed-integer linear program. We prove that any feasible solution guarantees energy safety and implement our approach in LLVM. Our evaluation on 13 benchmarks across 3 capacitor sizes shows that Bao outperforms existing baselines, achieving 10% faster execution and 52% fewer region boundary hits on average compared to the best baseline.

This work proposes a data-driven framework that synthesizes output-feedback control for linear time-invariant multi-agent systems (MAS). We first develop two algorithms that use offline data samples of state and input trajectories to determine a Nash equilibrium (NE) for MAS. Next, we formulate a semidefinite program that, given the NE solution, computes the static output-feedback controllers for the agents in MAS. Compared to prior work, virtues of our framework include weaker controllability assumptions, the absence of online information and data-driven estimation of states. Next, we use the proposed framework to design control schemes for voltage and frequency regulation for nonlinear multi-nodal electrical networks with grid-forming and grid-following inverters. Thus, compared to prior work that implements their data-driven frameworks in small-scale numerical examples, we validate the performance of our data-driven framework in high-fidelity nonlinear models of MAS. This implementation in realistic MAS provides insights that small-scale numerical examples may fail to capture. Results show successful grid stabilization and provision of regulation services from the inverters in the grid using our proposed framework. It can attain up to 78% less overshoot and 30% faster settling time than traditional strategies such as proportional-integral control.

Evaluating the occurrence and triggers of large language model (LLM) behaviors - such as sycophancy, self-preference, or over-confidence - is critical for predicting real-world model deployment risks. However, existing situated behavioral evaluations typically ignore cultural context, limiting their generalizability across an increasingly global user base. To address this gap, we propose CuBEs - Culturally-situated Behavior Evaluations that probe for response patterns across diverse user cultures. We first extend an automated testing pipeline to inject cultural context into behavioral test scenarios and subsequent evaluation. We assess the cultural adaptability of this pipeline by building a human-labeled dataset that captures nuanced dimensions of behavior understanding across 12 distinct cultures. Our dataset reveals significant cross-cultural variations that one-size-fits all judgments fail to capture. Through evaluating 13 open- and closed-source LLMs, we find that introducing cultural situatedness in the evaluation scenario creates significant variation in the presence of a behavior. For example, while our baseline experiments testing for political bias capture localized Western political dimensions like the American conservative-progressive divide, non-Western culturally situated evaluations surface entirely different axes of bias such as religious and colonial political issues. Our findings demonstrate that standard, culturally-agnostic evaluations fail to capture these shifts, highlighting the necessity of culturally situated behavioral testing for global deployments.

We study a family of quantum subsystem codes defined on rectangular coordinate grids of any dimension, with even side lengths of at least four. The directly measured operators act along coordinate lines. For every member of the family, we derive the number of protected logical qubits and prove its dressed distance without a numerical search. To illustrate the resulting design choices, we impose a budget of at most 10,000 data qubits and determine the complete tradeoff between encoding rate and measurement weight within this family at distances 16, 32, and 64. At distance 16, one code protects 4,096 logical qubits among 10,000 data qubits using weight-ten measurements. At distances 32 and 64, codes with 7,776 and 9,216 data qubits protect 1,024 and 256 logical qubits, respectively, using weight-six measurements. We include and explain Lean code snippets throughout the paper to guide readers through the formal verification of our subsystem-code construction and its dressed and bare distances.

This paper is concerned with the effect of spatial heterogeneity on the linear stability of unique positive steady states for a nonlocal population equation. The main difficulty is that, because of the nonlocal term, the linearized operator need not be self-adjoint or satisfy the maximum principle, so the usual stability arguments based on the principal eigenvalue do not apply directly. We establish linear stability for sufficiently large diffusion and, under suitable conditions on the maximum set of the local growth ratio, for sufficiently small diffusion. In one space dimension, we also construct unstable positive steady states and obtain Hopf bifurcations for sufficiently small diffusion. Keeping the local coefficients fixed, we obtain either a supercritical or a subcritical Hopf bifurcation by changing the weight in the nonlocal integral. Our results show that uniqueness of a positive steady state does not imply its linear stability.

We measured the 198-310 GHz flux densities of 18 Class II protoplanetary disks in the Ophiuchus star-forming complex using the Submillimeter Array (SMA). The $\sim$200-300 GHz spectral indices ($\alpha_{200-300}$) of these disks are $2.3^{+0.2}_{-0.5}$. In this observed frequency range, emission properties of the Ophiuchus disks are very similar to the 48 spatially compact Class II disks in the Taurus-Auriga region observed in a previous SMA survey. Combining the Ophiuchus and Taurus-Auriga samples, the Pearson correlation coefficient between $\alpha_{200-300}$ and the flux density at 250 GHz ($F_{250\,\mathrm{GHz}}$) is 0.27 ($p=0.02$). We do not find a convincing correlation between $\alpha_{200-300}$ and the $R_{95\%}$ radius of the disks. Our interpretation for the $\alpha_{200-300}$ values being close to 2 in the spatially compact disks is that at $\gtrsim 200$ GHz frequencies, the continuum emission in these disks is dominated by optically thick dust thermal emission, independent of $F_{250\,\mathrm{GHz}}$ and $R_{95\%}$. As a result, the (sub)millimeter luminosity and the apparent disk size are positively correlated, which has been found in numerous resolved emission surveys of Class II disks. In young clusters with ages of $\sim$1-2 Myr, Class II disks with smaller radii or lower (sub)millimeter luminosities are not necessarily those in which the dust mass budget has been consumed or dispersed.

Generative and embodied AI services exchange token streams within continuing inference and control loops. Their communication requirements depend on each token set's purpose, useful timing, and execution context. This article organizes these properties into service, temporal, and stateful semantics and proposes a token service interface (TSI) between applications, radio access networks (RANs), and edge runtimes. TSI binds delivery requirements, readiness forecasts, and execution-state references to each schedulable token set, specifying field ownership, versioned updates, and admission feedback. A drone inspection case study over a decoupled RAN illustrates importance-aware radio allocation, advance preparation for timely delivery, and selective state migration that balances interruption against forwarding delay.