Artificial intelligence has shown promise in medical imaging, yet most existing systems lack flexibility, interpretability, and adaptability - challenges especially pronounced in ophthalmology, where diverse imaging modalities are essential. We present EyeAgent, the first agentic AI framework for comprehensive and interpretable clinical decision support in ophthalmology. Using a large language model (DeepSeek-V3) as its central reasoning engine, EyeAgent interprets user queries and dynamically orchestrates 53 validated ophthalmic tools across 23 imaging modalities for diverse tasks including classification, segmentation, detection, image/report generation, and quantitative analysis. Stepwise ablation analysis demonstrated a progressive improvement in diagnostic accuracy, rising from a baseline of 69.71% (using only 5 general tools) to 80.79% when the full suite of 53 specialized tools was integrated. In an expert rating study on 200 real-world clinical cases, EyeAgent achieved 93.7% tool selection accuracy and received expert ratings of more than 88% across accuracy, completeness, safety, reasoning, and interpretability. In human-AI collaboration, EyeAgent matched or exceeded the performance of senior ophthalmologists and, when used as an assistant, improved overall diagnostic accuracy by 18.51% and report quality scores by 19%, with the greatest benefit observed among junior ophthalmologists. These findings establish EyeAgent as a scalable and trustworthy AI framework for ophthalmology and provide a blueprint for modular, multimodal, and clinically aligned next-generation AI systems.

Motivated by intracellular phase separation, we theoretically investigate how molecular orientation and multi-component nature affect phase behavior. We construct a minimal model for a ternary mixture composed of isotropic (I), anisotropic (A), and solvent (s) components by combining the Flory-Huggins and Maier-Saupe theories. We obtain two main results from evaluating the phase behavior and the time evolution of the density fields. First, for certain interaction parameters, two distinct binodal lines appear in the plane of the volume fractions of the I- and A-components, and merge through their respective critical points. Second, rapid droplet formation emerges due to a weakly first-order phase transition, characterized by a discontinuity of the spinodal surface. The first result indicates the possibility of continuous transformation between the two phase-separated states. The second result suggests that anisotropic molecules can regulate phase separation kinetics. These findings might be physically general beyond biological systems.

Can classical game-theoretic frameworks be extended to capture the bounded rationality and causal reasoning of AI agents? We investigate this question by extending Causal Normal Form Games (CNFGs) to sequential settings, introducing Sequential Causal Multi-Agent Systems (S-CMAS) that incorporate Pearl's Causal Hierarchy across leader-follower interactions. While theoretically elegant -- we prove PSPACE-completeness, develop equilibrium refinements, and establish connections to signaling theory -- our comprehensive empirical investigation reveals a critical limitation: S-CNE provides zero welfare improvement over classical Stackelberg equilibrium across all tested scenarios. Through 50+ Monte Carlo simulations and hand-crafted synthetic examples, we demonstrate that backward induction with rational best-response eliminates any strategic advantage from causal layer distinctions. We construct a theoretical example illustrating conditions where benefits could emerge ($ε$-rational satisficing followers), though implementation confirms that even relaxed rationality assumptions prove insufficient when good instincts align with optimal play. This negative result provides valuable insight: classical game-theoretic extensions grounded in rational choice are fundamentally incompatible with causal reasoning advantages, motivating new theoretical frameworks beyond standard Nash equilibrium for agentic AI.

The Euler-Bernoulli beam model has been studied classically and semi-classically. The semi-classical quantization is done in an analogous way to the quantization of the electromagnetic field, and we found an effect that is similar to the Casimir effect, which is the photonic Casimir effect. The Casimir force, by unit area, is proportional to the first mode energy divided by the volume of the beam. For the hinged-hinged boundary condition, degenerate states were found. These degenerate pairs form decoherence-free subspaces for dispersive thermal reservoirs. For other boundary conditions, there are also subspaces with lower decoherence rates, which occur for quasi-degenerate states.

Algorithms for computing neutrino oscillation probabilities in sharply varying matter potentials such as the Earth are becoming increasingly important. As the next generation of experiments, DUNE and HyperK as well as the IceCube upgrade and KM3NeT, come online, the computational cost for atmospheric and solar neutrinos will continue to increase. To address these issues, we expand upon our previous algorithm for long-baseline calculations to efficiently handle probabilities through the Earth for atmospheric, nighttime solar, and supernova neutrinos. The algorithm is fast, flexible, and accurate. It can handle arbitrary Earth models with two different schemes for varying density profiles. We also provide a c++ implementation of the code called NuFast-Earth along with a detailed user manual. The code intelligently keeps track of repeated calculations and only recalculates what is needed on each successive call which can also help provide significant speed-ups.

Physical Gottesman-Kitaev-Preskill (GKP) states are inherently noisy as ideal ones would require infinite energy. While this is typically considered as a deficiency to be actively corrected, this work demonstrates that imperfect GKP stabilizer states can be leveraged in order to apply non-Clifford gates using only linear optical elements. In particular, Gaussian operations on normalizable GKP states, combined with homodyne measurements, permit two key primitives: clean projection onto Pauli eigenstates in the normalizable GKP codespace, thereby implementing Clifford gates with high fidelity; and probabilistic projection of unmeasured modes onto non-Pauli eigenstates. These results demonstrate that normalizable GKP stabilizer states combined with Gaussian operations provide a practical framework for computational universality within the measurement-based model of quantum computation in a realistic continuous-variable setting.

This work demonstrates that repeated weak measurements together with post-selection can produce sharp dynamical discontinuities in meter observables, even in minimal quantum systems. The discontinuous behavior is governed by the polar angle of the post selected state, which serves as a continuous control parameter. As this angle is varied, the expectation value of the meter observables changes abruptly at the point where the imaginary part of the associated weak value, a complex quantity that arises in weak measurements with post-selection, becomes zero. Such non-analytic behavior emerges only when the weak value is genuinely complex for a range of post-selection angles. If the weak value remains purely real for all angles, the dynamics remain smooth. The discontinuity originates from an exchange of stability between fixed points of the non-unitary Kraus operator governing the meter's evolution. Remarkably, despite the absence of a thermodynamic limit, the relaxation time in the vicinity of the discontinuity exhibit universal critical behavior characterized by a critical exponent equal to $1$, independent of system parameters. These results establish the weak value as a tunable control parameter capable of inducing non-analytic dynamical responses and reshaping the stability structure of measurement-induced quantum dynamics.

Estimating the fidelity between an unknown quantum state and a fixed target is a fundamental task in quantum information science. Direct fidelity estimation (DFE) enables this without full tomography by sampling observables according to a target-dependent distribution. However, existing approaches face notable trade-offs. Grouping-based DFE achieves strong accuracy for small systems but suffers from exponential scaling, and its applicability is restricted to Pauli measurements. In contrast, classical-shadow-based DFE offers scalability but yields lower accuracy on structured states. In this work, we address these limitations by developing two classes of operator-aware shadow importance sampling algorithms using informationally overcomplete positive operator-valued measures. Instantiated with local Pauli measurements, our algorithm improves upon the grouping-based algorithms for Haar-random states. For structured states such as the GHZ and W states, our algorithm also eliminates the exponential memory requirements of previous grouping-based methods. Numerical experiments confirm that our methods achieve state-of-the-art performance across Haar-random, GHZ, and W targets.

Stimulated Raman scattering (SRS) microscopy has emerged as a powerful technique for probing the spatiotemporal dynamics of molecular bonds with exceptional sensitivity, resolution, and speed. However, classically, its performance remains fundamentally constrained by optical shot noise, which imposes a strict limit on detection sensitivity and speed. Here, we demonstrate a quantum-enhanced SRS microscopy platform that circumvents this barrier by harnessing amplitude-squeezed light. Specifically, we generate a Stokes beam with $5.2~\mathrm{dB}$ of amplitude squeezing using traveling-wave optical parametric amplification in second-order nonlinear waveguides, and combine it with a tunable coherent pump to access vibrational modes spanning from $1000$ to $3100~\mathrm{cm}^{-1}$. Applied to quantum imaging of metabolites in biological tissue (pork muscle), our quantum-enhanced Raman microscope achieves an average noise suppression of $3.6~\mathrm{dB}$ and a $51\%$ enhancement in signal-to-noise ratio (SNR) -- to the best of our knowledge, the largest improvement reported to date in quantum-enhanced SRS microscopy of biological samples.

We show that the Skolem Problem is decidable in finitely generated commutative rings of positive characteristic. More precisely, we show that there exists an algorithm which, given a finite presentation of a (unitary) commutative ring $\mathcal{R} = \mathbb{Z}_{/T}[X_1, \ldots, X_n]/I$ of characteristic $T > 0$, and a linear recurrence sequence $(γ_n)_{n \in \mathbb{N}} \in \mathcal{R}^{\mathbb{N}}$, determines whether $(γ_n)_{n \in \mathbb{N}}$ contains a zero term. Our proof is based on two recent results: Dong and Shafrir (2026) on the solution set of S-unit equations over $p^e$-torsion modules, and Karimov, Luca, Nieuwveld, Ouaknine, and Worrell (2025) on solving linear equations over powers of two multiplicatively independent numbers. Our result implies, moreover, that the zero set of a linear recurrence sequence over a ring of characteristic $T = p_1^{e_1} \cdots p_k^{e_k}$ is effectively a finite union of $p_i$-normal sets in the sense of Derksen (2007).

We consider some boundary value tracking optimal control problem constrained by a Neumann boundary value problem for some elliptic partial differential equation where the control acts as right-hand side. This optimal control problem can be reformulated asa state-based variational problem that is the starting point for the finite element discretizion. In this paper, we only consider atensor-product finite element discretizion for which optimal discretization error estimates and fast solvers can be derived.Numerical experiments illustrate the theoretical results quantitatively.

As distributed energy resources (DERs) proliferate, future power system will need new market platforms enabling prosumers to trade various electricity and grid-support products. However, prosumers often exhibit complex, product interdependent preferences and face limited cognitive and computational resources, hindering engagement with complex market structures and bid formats. We address this challenge by introducing a multi-product market that allows prosumers to express complex preferences through an intuitive format, by fusing combinatorial clock exchange and machine learning (ML) techniques. The iterative mechanism only requires prosumers to report their preferred package of products at posted prices, eliminating the need for forecasting product prices or adhering to complex bid formats, while the ML-aided price discovery speeds up convergence. The linear pricing rule further enhances transparency and interpretability. Finally, numerical simulations demonstrate convergence to clearing prices in approximately 15 clock iterations.

A reliable calculation of radiative corrections to $τ\toππν_τ$ decays is an important prerequisite for using hadronic $τ$ decays for a data-driven evaluation of the hadronic-vacuum-polarization contribution to the anomalous magnetic moment of the muon, $a_μ^\text{HVP, LO}[ππ,τ]$. In this Letter, we present an improved model-independent analysis of these radiative corrections, including, for the first time, effects beyond point-like pions in the evaluation of the loop diagrams. These structure-dependent corrections, implemented via a dispersive representation of the pion form factor, lead to significant changes compared to previous calculations due to enhancements near the $ρ(770)$ resonance. We also devise strategies for the matching to chiral perturbation theory and a stable implementation of the real corrections down to the two-pion threshold, which shows that some higher-order isospin-breaking corrections need to be kept due to a strong threshold enhancement. Finally, we perform dispersive fits to the currently available $τ\toππν_τ$ spectra and discuss the consequences for isospin-breaking corrections in the evaluation of $a_μ^\text{HVP, LO}[ππ,τ]$.

Rayleigh-Benard convection (RBC) is a canonical system for buoyancy-driven turbulence and heat transport, central to geophysical and industrial flows. Developing efficient control strategies remains challenging at high Rayleigh numbers, where fully resolved simulations are computationally expensive. We use a control framework that couples data-driven manifold dynamics (DManD) with reinforcement learning (RL) to suppress convective heat transfer. We find a coordinate transformation to a low-dimensional system using POD and autoencoders, and then learn an evolution equation for this low-dimensional state using neural ODEs. The reduced model reproduces key system features while enabling rapid policy training. Policies trained in the DManD environment and deployed in DNS achieve a 16-23 % reduction in the Nusselt number for both single- and dual-boundary actuation. Physically, the learned strategy modulates near-wall heat flux to stabilize and thicken the thermal boundary layer, weaken plume ejection, and damp the wall-driven instabilities that seed convective bursts. Crucially, the controller drives the flow toward a quasi-steady state characterized by suppressed temporal fluctuations and spatially steady heat-flux patterns. This work establishes DManD-RL as a physically interpretable, scalable approach for turbulence control in high-dimensional flows.

Item IDs form the backbone of industrial recommender systems, but suffer from representation instability and poor long-tail generalization in large, dynamic item corpora. Semantic IDs (SIDs) mitigate these issues by enabling knowledge sharing through quantization of item content features. Existing methods attempt to enhance SID expressiveness by incorporating collaborative information with content features; however, they often overlook a critical distinction: unlike relatively uniform content features, user-item interactions are highly skewed, resulting in a significant quality gap in collaborative information between popular and long-tail items. This mismatch leads to two critical limitations: (1) Collaborative Noise Corrupts Behavior-Content Alignment: Behavior-content alignment is a prevailing approach for modeling shared information. However, indiscriminate alignment allows collaborative noise from long-tail items to corrupt their content representations, leading to the loss of critical multimodal information. (2) Collaborative Noise Obscures Critical Behavioral SIDs: When modeling modality-specific information, prior works typically generate multiple behavioral SIDs with equal weights for each item. This equal-weight scheme fails to reflect the varying importance of different behavioral SIDs, making it difficult for downstream tasks to distinguish informative SIDs from noisy ones. To address these challenges, we propose ADC-SID, a framework that Adaptively Denoises Collaborative information for SID quantization. It comprises two key components: (i) Adaptive Behavior-Content Alignment, which adjusts alignment strength to mitigate corruption caused by collaborative noise; and (ii) Dynamic Behavioral Weighting Mechanism, which learns importance scores for behavioral SIDs to enable downstream models to suppress noise. Extensive experiments has demonstrated ADC-SID's superiority...

The empirical Orlicz norm based on a random sample is defined as a natural estimator of the Orlicz norm of a univariate probability distribution. A law of large numbers is derived under minimal assumptions. The latter extends readily to a linear and a nonparametric regression model. Secondly, sufficient conditions for a central limit theorem with a standard rate of convergence are supplied. The conditions for the CLT exclude certain canonical examples, such as the empirical sub-Gaussian norm of normally distributed random variables. For the latter, we discover a nonstandard rate of $n^{1/4} \log(n)^{3/8}$, with a heavy-tailed, stable limit distribution. It is shown that in general, the empirical Orlicz norm does not admit any uniform rate of convergence for the class of distributions with bounded Orlicz norm.

Vaccine randomized trials are typically designed to be blinded, ensuring that the estimated vaccine efficacy (VE) reflects the immunological effect of the vaccine. When blinding is broken, however, the estimated VE reflects not only the immunological effect but also behavioral effects stemming from participants' awareness of their treatment status. Recent work has proposed alternative causal estimands to the standard VE to address this issue, but their point identification results require a strong assumption: the absence of unmeasured common causes of infection risk and participants' belief about whether they received the vaccine. Personality traits, for example, may plausibly violate this assumption. We relax this assumption and derive nonparametric causal bounds for different types of VE. We construct these bounds using two approaches: linear programming-based and monotonicity-based methods. We further consider several possible causal structures for vaccine trials and show how the nonparametric bounds differ across these scenarios. Finally, we illustrate the performance of the proposed bounds using fully synthetic data and a semi-synthetic data example based on a COVID-19 vaccine trial.

Videos of the 2020 Beirut explosion offer a rare opportunity to see a shock wave. We summarize the non-linear theory of a weak shock, derive the Landau-Whitham formula for the thickness of the overpressure layer and, using frame-by-frame video analysis, we demonstrate agreement of data and theory.

In complex networks there are overlapping substructures or "circles" that consist of nodes belonging to multiple cohesive subgroups. Yet the role of these overlapping nodes in influence spreading processes remains underexplored. In the present study, we analyse networks with circle structures using a probabilistic influence spreading model for processes of simple and complex contagion. We quantify the roles of nodes using three metrics, i.e., In-Centrality, Out-Centrality, and Betweenness Centrality that represent the susceptibility, spreading power, and mediatory role of nodes, respectively, and find that at each stage of the spreading process the overlapping nodes consistently exhibit greater influence than the non-overlapping ones. Furthermore, we observe that the criteria to define circles shape the overlapping effects. When we restrict our analysis to only largest circles, we find that circles reflect not only node-level attributes but also of topological importance. These findings clarify the distinction between local attribute-driven circles and global community structures, thus highlighting the strategic importanc of overlapping nodes in spreading dynamics. This provides foundation for future research on overlapping nodes in both circles and communities.

We report the first detection of gamma-ray emission up to ultra-high-energy (UHE; $>$100 TeV) emission from the prototypical gamma-ray binary system LS I +61 303 using data from the Large High Altitude Air Shower Observatory (LHAASO). It is detected with significances of 9.2$σ$ in WCDA (1.4--30.5 TeV) and 6.2$σ$ in KM2A (25--267 TeV); in KM2A alone we identify 16 photon-like events above 100 TeV against an estimated 5.1 background events, corresponding to a 3.8$σ$ detection. These results provide compelling evidence of extreme particle acceleration in LS I +61 303. Furthermore, we observe orbital modulation at 3.9$σ$ confidence level, between 25 and 100 TeV, with a hint that the orbital modulation is energy-dependent. These features can be understood in a composite scenario in which leptonic and hadronic processes jointly contribute.

We compute the electronic and emission properties of Coulomb-correlated multi-particle states (X$^0$, X$^\pm$, XX) in weakly confining GaAs/AlGaAs quantum dots using an 8-band $\mathbf{k}\!\cdot\!\mathbf{p}$ model coupled to continuum elasticity and configuration interaction (CI). We evaluate polarization-resolved oscillator strengths and radiative rates both in the dipole approximation (DA) and in a quasi-electrostatic beyond-dipole (BDA) longitudinal formulation implemented via a Poisson reformulation exactly equivalent to the dyadic Green-tensor kernel. For the dots studied, BDA yields lifetimes in quantitative agreement with experiment, e.g., $τ^X=0.279\,\mathrm{ns}$ vs $0.267\,\mathrm{ns}$ (exp.) and $τ^{XX}=0.101\,\mathrm{ns}$ vs $0.115\,\mathrm{ns}$ (exp.). The framework also reproduces electric-field tuning of the multi-particle electronic structure and emission-including the indistinguishability inferred from $P=τ^X/(τ^X+τ^{XX})$-and we assess sensitivity to CI-basis size and to electron-electron and hole-hole exchange.

We generalize the theory of stable canonical rules by adopting definable filtration, a generalization of the method of filtration. We show that for a modal rule system or a modal logic that admits definable filtration, each extension is axiomatizable by stable canonical rules. Moreover, we provide an algebraic presentation of Gabbay's filtration and generalize stable canonical formulas and the axiomatization results via stable canonical formulas for $\mathsf{K4}$ to pre-transitive logics $\mathsf{K4^{m+1}_{1}} = \mathsf{K} + \Diamond^{m+1} p \to \Diamond p$ $(m \geq 1)$. As consequences, we obtain the fmp of $\mathsf{K4^{m+1}_{1}}$-stable logics and a characterization of splitting and union-splitting logics in the lattice $\mathsf{NExt}\mathsf{K4^{m+1}_{1}}$. There are continuum many $\mathsf{K4^{m+1}_{1}}$-stable logics that are neither $\mathsf{K4}$-stable logics nor subframe logics. Finally, we introduce $m$-stable canonical formulas, strengthening the axiomatization results for these logics.

Deployment of optimization algorithms over communication networks face challenges associated with time delays and corruptions. Fixed time delays can destabilize popular gradient-based algorithms, and this degradation is exacerbated by time-varying delays that may arise from packet drops. This work concentrates on the analysis and synthesis of discrete-time optimization algorithms with certified exponential convergence rates that are robust against switched network dynamics between the optimizer and the gradient oracle. Analysis is accomplished by solving linear matrix inequalities under bisection in the exponential convergence rate, searching over Zames-Falb filter coefficients that can certify convergence. Synthesis is performed by alternating between a search over filter coefficient for a fixed controller, and a search over controllers for a fixed filter. Effectiveness is demonstrated by the synthesis of convergent optimization algorithms over networks with time-varying delays, and networks with unstable channel dynamics.

In this paper, we present the design and preliminary performance evaluation of a new heavy-ion detector for direct measurements of heavy Λ hypernuclei lifetime. The detector employs the previously developed 10 picosecond resolution Radio Frequency (RF) Timer, which converts the temporal information of incident particles into spatial coordinates of secondary or photoelectrons on a position-sensitive detector by means of circular RF scanning in the 500-1000 MHz range. Here, we report the detector design to achieve efficient suppression of accidental background and effective separation of prompt reaction products and delayed events from Λ hypernuclei decays, results of test studies carried out with RF synchronized laser as well as preliminary results obtained by using alpha particles. Dedicated Monte-Carlo simulations have been performed to estimate the detector's performance under realistic experimental conditions at RF-driven electron, photon, or proton beams. The results confirm the feasibility of the proposed design and provide a basis for upcoming experimental measurements, based on the delayed fission detection.

We consider Fractional Quantum Hall (FQH) edges with a spatially local Quantum Point Contact (QPC). Within the Unified Nonequilibrium Perturbative (UNEP) framework, without assumptions on the underlying Hamiltonian $H_{0}$ for the edges, we search for the associated backscattering DC current and noise compatible with the anyonic time exchange (ATE) constraint with a phase $\barθ$. For that, we infer a nonequilibrium fluctuation-dissipation relation that explicitly involves $\barθ$ and yields an integral equation connecting the nonequilibrium DC current and noise. On one hand, we assume initial thermal states, so that the DC noise is Poissonian. Then the integral equation for the DC current is shown, through the Wiener-Hopf technique, to admit the unique TLL local solution. Therefore, $\barθ$ is necessarily tied to the scaling dimension $δ$, which is robust with respect to edge interactions. On the other hand, we address the "anyon collider" setup where DC noise is super-Poissonian. As the difference between nonequilibrium and equilibrium correlators is fixed, the integral equation admits a unique solution for both nonequilibrium DC backscattering current and super-Poissonian noise, whose explicit temperature dependence is thus determined.

We extend the notions of higher Du Bois and higher rational singularities to pairs in the sense of the minimal model program. We extend numerous results to these higher pairs, including Bertini type theorems, stability under finite maps and that m-rational pairs are m-Du Bois. We prove these using a generalized Kovács-Schwede-type injectivity theorem for pairs, the main technical result of this paper.

In this letter, we provide a simple algorithm, anyPUB, to systematically derive the $2 \rightarrow 2$ scattering matrix in the high-energy limit for any kind of models, irrespective of their gauge group or their field representation. After computing the eigenvalues analytically and/or numerically from this matrix, we impose perturbative unitarity bounds on them. We tested our method on various models and validated the results against the literature. Finally, as a concrete application of our approach, we discuss the case of the minimal left-right symmetric model and derive, for the first time, the perturbative unitarity constraints in the Pati-Salam model.

Public opinion governance in social networks is critical for public health campaigns, political elections, and commercial marketing. In this paper, we addresse the problem of maximizing overall opinion in social networks by strategically modifying the internal opinions of key nodes. Traditional matrix inversion methods suffer from prohibitively high computational costs, prompting us to propose two efficient sampling-based algorithms. Furthermore, we develop a deterministic asynchronous algorithm that exactly identifies the optimal set of nodes through asynchronous update operations and progressive refinement, ensuring both efficiency and precision. Extensive experiments on real-world datasets demonstrate that our methods outperform baseline approaches. Notably, our asynchronous algorithm delivers exceptional efficiency and accuracy across all scenarios, even in networks with tens of millions of nodes.

An Euler-type framework with equidistant step sizes is proposed for a class of time-changed stochastic differential equations.We establish the strong convergence rate of the standard Euler--Maruyama method under the global Lipschitz condition.The theoretical analysis is then extended to the truncated Euler--Maruyama method, proving its strong convergence under relaxed Khasminskii-type conditions.For both numerical schemes, the strong convergence orders are explicitly shown to be close to $α/2$, where $α\in (0,1)$ is the parameter of the time-change process.These results are significantly different from existing works using random step sizes, which typically preserve the classical convergence order of $1/2$.Numerical simulations are provided to demonstrate the theoretical findings.

We examine the effects of symmetry restoration on nuclear beta decay within the axially deformed proton-neutron quasiparticle random phase approximation (QRPA). We employ the proton-neutron finite-amplitude method (pnFAM) to compute transition amplitudes, and perform angular-momentum projection both after variation and after the QRPA to restore rotational symmetry. Exact projection reduces the calculated beta decay half-lives from those that use the needle approximation by up to 60%, and even more when taking the effects of projection on the ground-state energy into account.