For a sequence of binary bets, the Kelly criterion provides a closed-form solution that maximizes the expected growth rate of wealth. In contrast, when multiple bets are placed simultaneously (e.g., in portfolio allocation or prediction markets), the optimal Kelly strategy generally requires numerical optimization over a joint outcome space. A naive formulation scales exponentially in the number of bets, requiring $O(2^N)$ time and memory for $N$ simultaneous wagers, which restricts existing methods to small problem sizes. We present two complementary methods that dramatically extend the scale of multivariate Kelly problems that can be solved. First, in the case of independent bets, we introduce an integral transform formulation that eliminates explicit enumeration of outcomes, reducing the computational complexity of evaluating the objective from $O(2^N)$ to $O(N)$. Combined with numerically stable quadrature, this enables accurate solutions for problems involving hundreds of bets. Second, we develop a decomposition-based approach that constructs and solves carefully chosen subproblems, yielding feasible lower bounds and infeasible upper bounds on the optimal growth rate. This provides a practical mechanism for quantifying worst-case suboptimality as a function of subproblem size. Together, these methods make it possible to study the large-$N$ regime of the multivariate Kelly problem. Using synthetic data inspired by prediction markets, we show that the relationship between subproblem size and solution accuracy follows a simple and highly regular scaling law. In particular, the shortfall ratio between the lower and upper bounds is well-approximated by a sigmoid function of the relative subproblem size, with parameters that can be predicted from low-dimensional summary statistics of the problem.

We describe the public release of $>2.3$ petabytes of data from the FLAMINGO cosmological simulations. The suite consists of hydrodynamical simulations that include radiative cooling, star formation, stellar mass loss and the resulting chemical enrichment, supernova feedback, and two implementations of AGN feedback. Neutrinos are simulated explicitly using particles. Data products include snapshots, halo and galaxy catalogues, HEALPix all-sky lightcone maps, particle data for lightcone maps, and power spectra. The FLAMINGO set includes 22 hydrodynamical simulations. In addition, there are 16 gravity-only simulations, including the $10080^3$ particles FLAMINGO-10k run, with initial conditions that match those of the corresponding hydrodynamical runs. The fiducial hydrodynamical simulations span three numerical resolutions that have each been calibrated to reproduce the present-day galaxy stellar mass function and gas fractions in low-redshift clusters. Other simulations systematically vary the galaxy stellar mass function, cluster gas fractions, cosmology (including neutrino masses), and/or the nature of dark matter, in volumes of 1Gpc$^3$. The release includes hitherto unpublished simulations that use extra dark matter particles. While we provide a facility for downloading complete simulation outputs, we recognise that for many users this will not be possible due to limited local storage or network bandwidth. We implement a web service that enables users to explore available outputs and selectively download datasets or parts of datasets.

The temperature and magnetic field dependent magnetic penetration depth, $λ_m(T,H)$, was measured in a single crystal of a heavy fermion superconductor CeCoIn$_5$ using a frequency-domain tunnel diode resonator. In addition to the London penetration depth, which yields the superfluid density, measurements in a finite DC magnetic field provide Campbell penetration depth, $λ_C(T,H)$, which is directly linked to the true (unrelaxed) critical current density, $J_c$. The measured $λ_C(H)$ in CeCoIn$_5$ deviates significantly from the conventional $\sim \sqrt{H}$ behavior, and its slope changes abruptly at the characteristic magnetic field values. Considering that our sample is in the clean limit, we interpret this deviation as a fingerprint of the vortex lattice symmetry change. The temperature dependence $J_c(T)$ of CeCoIn$_5$ calculated from $λ_C(T)$ is nearly $T$-linear over the entire temperature range, also in stark contrast to expectations in a conventional type-II superconductor. Our results provide new evidence for unconventional superconductivity in CeCoIn$_5$ from the never-before-measured Campbell penetration depth.

We investigate whether Scalar-Tensor-Vector Gravity in its weak-field modified gravity form can account for the cluster-scale inverse-square force law inferred from recent kinematic Sunyaev-Zeldovich measurements of cluster pairwise motions. The starting point is the X-COP cluster fit of STVG-MOG, for which a representative baryonic cluster mass $M\sim 10^{15}M_\odot$ together with parameters $α\sim 9.11$ and $μ\sim 0.196~{\rm Mpc}^{-1}$ provides a successful description of cluster dynamics without particle dark matter. We extrapolate this fit to the separation range $30$ to $230~{\rm Mpc}$, relevant for the pairwise kSZ analysis. Since the Yukawa transition length $μ^{-1}\simeq 5.1~{\rm Mpc}$ is much smaller than these separations, the STVG-MOG acceleration law reduces to an effective inverse-square form. This explains why the theory can satisfy the observed Newtonian behavior while remaining distinct from MOND-like long-distance modifications. We derive the corresponding pairwise velocity curve and show that, after fitting a single overall kSZ amplitude, the extrapolated STVG-MOG prediction reproduces the measured trend of the pairwise kSZ data. The analysis shows that the X-COP cluster fit and the cosmological-scale kSZ force-law result are mutually consistent within STVG-MOG.

We develop a full-covariance formalism for pulsar timing array(PTA) -- astrometry verlap reduction function (ORF) estimators and use it to forecast graviton-mass constraints from a nanohertz stochastic gravitational-wave background (SGWB). Analytic covariance expressions are derived for auto- and cross-channel ORF estimators, including signal-signal, noise-noise, and signal-noise contributions, and are validated against numerical simulations. For an observational configuration with sensitivities comparable to NANOGrav and Gaia, we obtain an expected joint 90\% upper limit of $m_g<4.41\times10^{-24}\,\mathrm{eV}/c^2$, which remains PTA-dominated and lies at the same order of magnitude as the existing NANOGrav 15-year PTA-only bound. For a future-like configuration with sensitivities comparable to the SKA and Theia/Gaia-NIR, the astrometric channels contribute significantly to the constraining power, and the joint limit improves to $m_g<0.48 \times 10^{-24} \, \mathrm{eV}/c^2$. These forecasts indicate that PTA -- astrometry multichannel inference provides a viable avenue for improving graviton-mass constraints under next-generation observational conditions.

For $G$ a finite group, we show that the irreducible quandle representations of the conjugacy quandle $Conj(G)$ are the products of what we call characters of $Conj(G)$ by irreducible linear group representations of $G$. The unitary quandle representations can be distinguished. We apply this to the study of the enveloping group of $Conj(G)$ showing that it injects into $G \times \mathbb{Z}^{c_G}$, where $c_G$ is the number of conjugacy classes of $G$. The injection is an isomorphism for $G$ perfect.

An operator $T$ in a separable factor $\mathcal{M}$ is said to be irreducible in $\mathcal{M}$ if the von Neumann subalgebra $W^*(T)$ generated by $T$ is an irreducible subfactor of $\mathcal{M}$, i.e., $W^*(T)'\cap\mathcal{M}=\mathbb{C}I$. We say that $T$ is a single generator of $\mathcal{M}$ if $W^*(T)=\mathcal{M}$. In this paper, we study generators of separable factors related to maximal abelian self-adjoint subalgebras. As an application, we obtain a complete characterization of normal operators in separable factors which are similar to irreducible operators.

Modern distribution grids with high penetration of renewable generation provide substantial flexibility through distributed reactive power sources and transformer tap changers. This high degree of freedom can be exploited for optimisation. However, choosing an objective function for optimisation is not trivial, e.~g. minimising grid losses may lead to overvoltages and minimising voltage deviations may lead to higher reactive power flows to neighbouring system operators. Thus, this paper deals with the design of an objective function for the centralised optimisation of distributed reactive power sources and transformer tap changers. Different objectives for characteristic network quantities are investigated for the optimisation and optimised in a combined manner and separately. The consequences of optimising conflicting target values are then analysed. For the optimisation, various grid usage cases of a 110 kV benchmark power grid from SimBench are examined. The investigated power grid is characterised by a high proportion of renewable energy plants. The optimisation is carried out in a data-driven, object-oriented manner using the interior point method with open source software. At the end of the paper, a meaningful optimisation function with combined weighted objectives is derived and the results are analysed.

Various lower bounds are established for the entropy of sums, products and their combinations. First, we derive a prime-field analogue of a version of the entropy power inequality established by Tao over torsion-free groups. Next, we prove an entropy sum-product statement: For independent and identically distributed random variables $X,X'$, the maximum of ${\bf H}(X+X')$ and ${\bf H}(XX')$ is bounded below by a linear combination of the entropy and the min-entropy (Rényi entropy of order~$\infty$) of $X$. This result, obtained by bounding entropies of the form ${\bf H}\bigl( X(Y+Z)\bigr)$ from above and below, is valid over arbitrary fields $F$. Over $F={\bf R}$, a slightly stronger inequality is derived. Finally, a weak version of a purely Shannon-entropic sum-product result is developed: If the entropic additive doubling of a random variable $X$ over an arbitrary field is $O(1)$, then its multiplicative doubling is at least proportional to ${\bf H}(X)$.

In this Letter we explore the modelling of hadron production in electromagnetic ion dissociation (EMD) processes in high-energy ultraperipheral collisions at LHC energies. Since EMD can accompany exclusive particle production in these interactions, we demonstrate that the resulting hadrons can break the exclusivity vetos typically imposed by experiments. As two representative examples, we calculate the impact on existing LHC measurements of exclusive muon pair production ($γγ\toμμ$) and exclusive coherent $J/ψ$ production. We demonstrate that accounting for this effect resolves long-standing tensions between theoretical predictions and experimental measurements.

We develop an algebraic framework over arbitrary quadratic fields $L = \mathbb{Q}(\sqrt{D})$ to generalize the Miller-Rabin primality test. Consequently, we present a deterministic primality test for integers of the form $N = K p^{\ell} - 1$ that requires only a single modular exponentiation and achieves a computational complexity of $\tilde{\mathcal{O}}(\log^2 N)$. Furthermore, we also establish an analogue of Korselt's criterion within this setting. Finally, computational data generated using SageMath confirm its efficiency, successfully establishing the primality of numbers in the associated quadratic field within milliseconds.

We are concerned with the market-consistent valuation of lifelong health insurance products, which are subject to adjustments derived from the actuarial equivalence principle and driven by (medical) inflation. Such products are well-established in the European national markets, and the dynamics of the adjustment mechanism is well-understood from an actuarial perspective. However, the question of market-consistent valuation (as is necessary for Solvency II reporting) has not previously been addressed. This gap has led to a situation where some practitioners use stochastic models while others rely on deterministic methods to assign market-consistent values (Best Estimates) to the same type of health insurance liabilities. The purpose of this note is to fill this gap by showing that the Best Estimate of a lifelong health insurance policy depends on the choice of model for the interest and inflation rates. That is, the Best Estimate is not uniquely determined by the currently prevailing term structures of nominal and real spot rates, whence a deterministic calculation is theoretically unjustified. Furthermore, we construct a valuation portfolio such that the Best Estimate valuation decouples into calculations of 1.) deterministic coefficients derived from policy data and 2.) the prices of basis financial instruments that are independent of the individual policy data. Using this decomposition, the policies do not have to be tracked individually along each generated stochastic path. This allows for a more efficient evaluation of the Best Estimate for a large stock of policies with a stochastic model.

We propose a stochastic nonconvex optimization algorithm that achieves almost sure $\tilde{\mathcal{O}}(ε^{-1.5})$ iteration complexity for problems with smooth objective functions and gradients only observable with noise. The mean-zero stochastic noise is decision-dependent and has unbounded support with subexponential tail, allowing our framework to cover a broad class of problems. The improved almost sure iteration complexity is achieved with a new variant of the adaptive sampling trust-region optimization (ASTRO) augmented with an adaptively regularized local model, which we term Reg-ASTRO. Adaptive sampling ensures that the estimation precision is aligned with a measure of stationarity, so that iterates closer to stationarity trigger higher accuracy requirement for sampling. A key analytical challenge arises because the trust-region radius and regularization are coupled and not determined prior to gradient estimation at each iteration. We further establish an almost sure $\tilde{\mathcal{O}}(ε^{-4.5})$ sample complexity for Reg-ASTRO, which improves to $\tilde{\mathcal{O}}(ε^{-3.5})$ under stronger regularity conditions and use of common random numbers, substantially outperforming first-order methods in theory and numerical experiments.

We predict ground-state energies of $^3$H, $^4$He, and $^6$Li in chiral effective field theory up to next-to-next-to-next-to-leading-order (N$^3$LO) using a power counting guided by renormalization-group invariance. Subleading two-nucleon interactions are treated perturbatively, and for $^4$He and $^6$Li, we calculate the perturbative corrections from numerical derivatives of ground-state energies obtained with Lanczos diagonalization. We find that including the $^3$H binding energy in the calibration is essential for robust predictions of $^4$He and $^6$Li. This work demonstrates that the employed power counting can be applied to construct nuclear interactions with predictive power for light nuclei, bringing nuclear structure predictions closer to a foundation in quantum chromodynamics.

Materials discovery is fundamental to advance next-generation technologies as well as for sustainable and circular economy. Beyond computational screening, generative models are efficient at finding materials with desired properties, via multi-modal learning using multiscale data. This perspective examines the landscape of generative design for inorganic materials and discusses the integration of multi-modal learning with high-throughput experimental validation. We contextualize these challenges through the lens of a generative design framework as a unified approach to address the data-driven inverse design of functional materials. The central idea of the framework is constructed around a foundation AI model for inorganic materials interlinked deeply with various property databases and high-throughput experiments via a machine learning driven closed loop, which enables the framework to solve key challenges in functional materials. We argue that domain-specific implementations of such integrated workflows represent a promising pathway toward the unresolved challenge of data-driven inverse design for atom-engineered inorganic functional materials.

In the past few decades, large universal structures have been found that challenge the homogeneity and isotropy expected in standard cosmological models. This study examines burst clustering in both galactic hemispheres using a recently developed methodology, using spheres in 3D space for testing regularities. Using our new method in both hemisphere we find only one deviation from isotropy. A small one in the Southern and a huge one in the Northern hemisphere. This itself suggests that the two deviations do not likely to come from statistical fluctuation. The northern huge group contains app. 125 gamma-ray bursts (GRBs) corresponds with the so - called Hercules-Corona Borealis Great Wall. The southern group contains 4-5 GRBs locating very close to each other. Two of them (GRB050822 and GRB050318) are close not just in redshift and the angular position but they are very close in observing time (5 months). We concluded that the third important result of this work that this method could not find other overdensity deviation from homogeneity in the GRBs spatial distribution. We have shown that the large-scale density increase in the spatial distribution of gamma-ray bursts does not necessarily violate the cosmological principle.

We consider the energy-optimal control problem for double-integrator systems subject to state and control constraints, with fixed terminal time and free terminal speed. When the constraints become active, the optimal trajectory consists of a combination of bang, unconstrained, and coast arcs, whose switching instants must be computed explicitly. In this paper, we derive closed-form expressions for the switching times of all admissible profiles, including both constrained and unconstrained arcs, reducing the computation in each case to explicit algebraic equations. In contrast to prior work, we classify all possible combinations of arcs, including special cases, and provide the specific conditions under which each case arises. Furthermore, we prove that when the initial unconstrained trajectory violates both speed and control constraints, the optimal solution follows a predetermined bang-affine-coast profile, enabling direct identification of the optimal trajectory without intermediate feasibility checks.

We establish rigidity results for holomorphic mappings and plurisubharmonic functions in complex geometry. First, under mild conditions, we show that the gradient of a $\operatorname{U}(1)$-invariant strictly plurisubharmonic function in $\mathbb{C}^2$ possesses finite fibers and induces a analytic mapping of topological degree $1$ on the symplectic quotient. Second, we prove that continuous fiber-wise holomorphic maps on proper fibrations elevate to global holomorphic maps when anchored by mutually disjoint sections, yielding rigidity for homomorphisms between elliptic fibrations and Abelian schemes. Third, we demonstrate that a fiber-wise holomorphic map of mapping degree $1$ from a fibered compact Kobayashi hyperbolic manifold to a projective variety is a biholomorphism, provided it is injective on a very ample hypersurface. Finally, we prove that a holomorphic Lie group action with sufficiently large orbits confines the critical locus of a proper invariant strictly plurisubharmonic function to the fixed-point set, guaranteeing a unique global minimum and yielding a sharp differential topological obstruction on the orbit dimensions of compact Lie group actions.

Dominance is a fundamental concept in compilers based on static single assignment (SSA) form. It underpins a wide range of analyses and transformations and defines a core property of SSA: every use must be dominated by its definition. We argue that this reliance on dominance has become increasingly problematic -- both in terms of precision and applicability to modern higher-order languages. First, control flow overapproximates data flow, which makes dominance-based analyses inherently imprecise. Second, dominance is well-defined only for first-order control-flow graphs (CFGs). More critically, higher-order programs violate the assumptions underlying SSA and classic CFGs: without an explicit CFG, the very notion that all uses of a variable must be dominated by its definition loses meaning. We propose an alternative foundation based on free variables. In this view, $ϕ$-functions and function parameters directly express data dependencies, enabling analyses traditionally built on dominance while improving precision and naturally extending to higher-order programs. We further present an efficient technique for maintaining free-variable sets in a mutable intermediate representation (IR). For analyses requiring additional structure, we introduce the nesting tree -- a relaxed analogue of the dominator tree constructed from variable dependencies rather than control flow. Our benchmarks demonstrate that the algorithms and data structures presented in this paper scale log-linearly with program size in practice.

As electric vehicles (EVs) are increasingly adopted as platforms for connected and automated vehicles (CAVs), enhancing their energy efficiency becomes critical. With the emergence of vehicle-to-vehicle (V2V) communication, cooperative adaptive cruise control (CACC) offers improved traffic flow, safety, and energy efficiency by enabling real-time coordination among EVs. However, conventional CACC algorithms neglected acceleration and regenerative braking dynamics in their implementation. To address this gap, this paper proposes a third-order dynamic model for EVs which has been derived from real-world experimental data. We also propose a novel, practical, and energy-efficient Lyapunov-based CACC controller explicitly designed for EV platoons. The proposed controller is requiring lower control gains while ensuring string stability and energy efficiency. To validate its effectiveness, we conduct both simulation and experimental environments, demonstrating that our approach reduces velocity fluctuations, maintains string stability at lower headway times, and improves energy efficiency of the CACC platoon by up to 38.5% compared to a baseline CACC.

We formulate a global-position colored-permutation encoding for the capacitated vehicle routing problem. Each of the $K$ vehicles selects a disjoint partial permutation, and the sum of these $K$ color layers forms a full $n\times n$ permutation matrix that assigns every customer to exactly one visit position. This representation uses $n^2K$ binary decision variables arranged as $K$ color layers over a common permutation structure, while vehicle capacities are enforced by weighted sums over the entries of each color class, requiring no explicit load register and hence no extra logical qubits beyond the routing variables. In contrast, many prior quantum encodings introduce an explicit capacity or load representation with additional qubits. Our construction is designed to exploit the Constraint-Enhanced QAOA framework together with its encoded-manifold analyses. Building on a requirements-based view of quantum utility in CVRP, we develop a routing optimization formulation that directly targets one of the main near-term bottlenecks, namely the additional logical-qubit cost of vehicle labels and explicit capacity constraints. Our proposal shows strong algorithmic performance in addition to qubit efficiency. On a standard benchmark suite, our end-to-end pipeline recovers the independently verified optima. The feasibility oracle may also be of independent interest as a reusable polynomial-time decoding and certification primitive for quantum and quantum-inspired routing pipelines.

The Chow polynomial of a matroid is a fundamental invariant whose coefficients exhibit strong positivity properties, including $γ$-positivity. We interpret the normalized Chow coefficients as a probability distribution and establish new inequalities for its central moments. As consequences, we obtain bounds on the number of flags of flats and inequalities on the roots of the Chow polynomial. We further relate these moment inequalities to algebraic geometry via the Hirzebruch $χ_y$-genus. This yields new inequalities for matroidal Chern numbers. In particular, for any matroid of rank $d+1$, we prove that $c_1c_{d-1}\le c_d$, with equality if and only if $d=1$ or the simplification of the matroid is Boolean.

In this work, we present a deep learning-based automatic multitrack music mixing system catered towards live performances. In a live performance, channels are often corrupted with acoustic bleeds of co-located instruments. Moreover, audio-visual synchronization is of critical importance thus putting a tight constraint on the audio latency. In this work we primarily tackle these two challenges of handling bleeds in the input channels to produce the music mix with zero latency. Although there have been several developments in the field of automatic music mixing in recent times, most or all previous works focus on offline production for isolated instrument signals and to the best of our knowledge, this is the first end-to-end deep learning system developed for live music performances. Our proposed system currently predicts mono gains for a multitrack input, but its design along with the precedent set in past works, allows for easy adaptation to future work of predicting other relevant music mixing parameters.

We calculate the photon-energy spectrum of the inclusive radiative decay $B\to X_sγ$, induced by the electromagnetic dipole operator $O_7$, to next-to-next-to-next-to-leading order and consider the complete corrections for light fermions, for the contributions with two closed massive fermion loops, and for the limit of large QCD colour factors $N_{\rm c}$ in the remaining part. We discuss the total decay rate both without and with a cut on the photon energy. In addition to the on-shell renormalization of the bottom-quark mass, we also consider the kinetic mass and the MSR mass schemes. The latter two lead to an improved perturbative behaviour of the decay rate.

In this letter, we present fast readout of Pauli spin blockade phenomena and interdot coupling tunability in a silicon double quantum dot (DQD) fabricated using industry-compatible processes. The interdot couplings are tuned with a second self-aligned gate layer. The charge sensing and spin readout are performed by using gate-based reflectometry techniques. The results pave the way for scalable fast readout of large-scale industry-standard manufactured Si spin qubit arrays.

A graph $G$ is $(a,b)$-sparse if every nonempty subgraph $H$ satisfies $e(H) \leq a v(H) - b$. We are interested in the conditions under which an $(a,b)$-sparse graph can be partitioned $E(G) = E(G_1) \cup E(G_2)$ such that for $i \in \{1,2\}$ we have that $G_i$ is $(a_i, b_i)$-sparse. Kuperwasser, Samotij, and Wigderson conjectured that a $(m,0)$-sparse graph can be partitioned into a $(1,1)$-sparse graph and a $(m,2m-1)$-sparse graph. We prove the conjecture in full. The Kohayakawa--Kreuter Conjecture for Families claims that $n^{-1/m_2}$ is the threshold function for the random graph being Ramsey a.a.s. for graph families $\mathcal{H}_1, \ldots \mathcal{H}_r$. Kuperwasser, Samotij, and Wigderson motivated their conjecture by proving that it is sufficient to establish the Kohayakawa--Kreuter Conjecture for Families.

Agent-native social platforms such as Moltbook are rapidly emerging, yet they inherit and amplify classical influence and abuse attacks, where coordinated agents strategically comment and upvote to manipulate visibility and propagate narratives across communities. However, rigorous measurement and learning-based monitoring remain constrained by the absence of longitudinal, graph-native datasets for agentic social networks that jointly capture heterogeneous interactions, temporal drift, and visibility signals needed to connect coordination behavior to downstream exposure. We introduce MoltGraph as a realistic longitudinal agentic social-network graph dataset for studying how agents behave, coordinate, and evolve in the wild, enabling reproducible measurement on emerging multi-agent social ecosystems. Using MoltGraph, we provide the first graph-centric characterization of Moltbook as a dynamic network: (i) heavy-tailed connectivity with power-law exponents in the range alpha in [1.86, 2.72], (ii) accelerating hub formation and attention centralization where the top 1% agents account for 29.00% of engagements, (iii) bursty, short-lived coordination episodes, 98.33% last under 24 hours, and (iv) measurable exposure effects across submolts. In matched analyses, posts receiving coordinated engagement exhibit 506.35% higher early interaction rates (within H=5 days) and 242.63% higher downstream exposure in feeds than non-coordinated controls.

Atomic Layer Deposition (ALD) process development is often hindered by time-consuming and precursor-intensive tuning cycles required to identify saturation conditions. We introduce a physics-informed Bayesian active learning (BAL) framework that autonomously tunes precursor pulse times by integrating a Langmuir adsorption model directly into the Gaussian Process (GP) kernel. A key innovation is a two-stage parameter estimation strategy that decouples noise filtering from physical parameter extraction: the GP first smooths noisy data through standard prediction, then Langmuir parameters are fitted to the noise-filtered GP predictions. This approach effectively separates signal from experimental noise. We evaluate the framework against a standard data-driven GP across four simulated regimes, demonstrating convergence within five iterations, up to fourfold improvement in prediction accuracy, and two to fourfold reduction in precursor usage. Experimental validation using TiO2 deposition via Tetrakisdimethylamido Titanium (TDMAT) and ozone confirms that the physics-informed model accurately identifies saturation times for high-coverage targets ($\geq$95\%), with observed deviations at lower saturation levels providing valuable insight into non-ideal desorption behaviors.

We study robust nonlinear filtering for stochastic models driven by Lévy processes, where the signal and observation processes are coupled through common Brownian and jump noise. Robustness, defined as the continuous dependence of the filter on the observation path, is essential whenever the observation process deviates from the idealized model, for instance when a path must be reconstructed from discrete-time samples. This question is well understood for continuous semimartingale systems but largely open in the presence of jumps. We construct a version of the filter and establish its continuity in two regimes. For processes with finitely many jumps on compact intervals, we prove continuity in both the rough $p$-variation and $p$-variation topologies on cadlag path space, without requiring a separability condition on the jump coefficients. For processes with infinitely many jumps, we prove continuity in a modified rough $p$-variation topology adapted to cadlag geometric rough paths, under an additional separability assumption. In both cases, our approach relies on Stratonovich and Marcus flow decompositions rather than the Itô-based methods of recent work. The resulting geometric rough-path lifts yield pathwise convergence guarantees and can be constructed directly from discrete observations without knowledge of the underlying probability law.

We present a theoretical and numerical study of resonant surface-plasmon (SP) excitation driven by the beating of two co-propagating laser pulses on a smooth cylindrical plasma-vacuum interface. Analytical expressions for the SP dispersion relation, field amplitude, geometric coupling factor, and resonance conditions are derived and validated by fully three-dimensional particle-in-cell simulations. We reveal that curvature-induced geometric effects can substantially modify the SP dispersion and enable resonant matching by laser beat waves. This is inaccessible in planar geometries or with a single laser. Under matched resonance conditions, a high-amplitude SP-based wakefield can be generated by a few gigawatt lasers, placing this mechanism within reach of state-of-the-art fibre lasers. It therefore opens a route toward portable laser-driven plasma wakefield accelerators.