Papers

Scaling has become a primary driver of progress in language and vision foundation models, yet its role in precise correspondence matching remains underexplored. In this work, we present Flow Any Scene Transformer (FAST), a scalable correspondence model driven by two key insights. First, we reveal that the query-key projections inside single-view vision foundation models encode a coarse yet reusable prior for cross-view matching. Second, reusing these pretrained projections in cross-attention form yields a highly effective initialization for a ViT-based matcher built from a single-view encoder. Guided by these insights, we build FAST upon a vanilla single-view foundation model, utilizing a zero-parameter rewiring strategy to convert selected self-attention layers into cross-attention for cross-view interaction. This design allows ViT-based matchers to scale with advances in single-view foundation models, bypassing the need for a dedicated pair-centric pretraining stage. To fully unlock the scaling potential of this formulation, we assemble a 6-million-pair training corpus for general-purpose dense 2D displacement estimation across diverse co-visible image pairs. Extensive experiments demonstrate that FAST achieves state-of-the-art performance across a wide range of benchmarks, while scaling favorably with both backbone size and training data.

We study the noise threshold for positive quantum capacity for the qubit depolarizing channel. We explore analytically the action of the qubit depolarizing channel on the symmetric subspaces of the input qubits, in the limit of asymptotically many uses of the channel. We observe the emergence of a bosonic Gaussian channel. Furthermore, the codes previously developed for the depolarizing channel can be translated to codes for the emergent Gaussian channel, and it is easier to further optimize these codes for the simpler emergent Gaussian channel. Translating these codes back to the depolarizing channel leads to extremely good input states for the coherent information of the depolarizing channel producing new lower bounds on the noise threshold for positive capacity. In addition to improved lower bounds on the threshold, this newly found link between depolarizing noise and Gaussian channels offers a novel perspective contributing to our understanding of these symmetric codes.

The mixing-plane method plays a pivotal role in steady simulations of multiple turbomachinery components. With advances in computational power, a whole-engine gas turbine simulation is no longer an elusive approach. However, whole-engine simulations require simultaneous coupling of the compressor, turbine and combustor. A key challenge is that the working fluid after the combustor can no longer be treated as a perfect gas, since the combustion introduces composition variations and combustion products. Conventional mixing-plane methods based on a fixed-composition gas cannot guarantee a thermodynamically consistent mixed-out state. To overcome the difficulty, this paper extends the classic mixing-plane approach to handle compressible reacting flows. The nonlinearity in the thermodynamic closure is addressed using a novel nested algorithm combining an outer pressure-root search with an inner enthalpy inversion. Physical state checks and power-law pressure sampling are incorporated to improve numerical robustness at low normal velocities where the pressure root approaches its upper physical bound. The performance of the method is demonstrated in three configurations: a quasi-1D interface test elucidating the thermodynamic stiffness of the proposed mixing-plane formulation for different fuel types and flow conditions; the Darmstadt transonic compressor case that verifies the proposed method reduces to the classic mixing-plane formulation when the mixture fraction reduces to zero; and a whole-engine simulation of the KJ66 micro-turbojet is used to demonstrate the performance of the method for multiphysics turbomachinery simulations. Good agreement with experimental data is observed, and the maximum relative mass-flow error across the mixing planes remains below 0.11%, demonstrating conservative and thermodynamically consistent interface coupling for compressible reacting turbomachinery flows.

The Finitistic Dimension Conjecture is one of the most important homological conjectures in the representation theory of finite-dimensional algebras. Recently, Gélinas arXiv:2004.04828 proposed the notion of "delooping level" to study the finitistic dimension Conjecture. This article reports a misconception between the delooping level and the tensor product of algebras. Namely, for finite-dimensional algebras $A$ and $B$, the difference $\operatorname{dell}(A\otimes_{\mathbf{k}}B) - \left(\operatorname{dell}(A)+ \operatorname{dell}(B)\right)$ can be arbitrarily large or negative.

Real nodal lines (RNLs) featuring real Chern numbers and second-order boundary modes have attracted widespread attention. In all previously reported realizations, an RNL is linked by another nodal line, and whether an RNL can be linked by other types of band degeneracies has remained open. Here, we propose a second-order real nodal-line semimetal in which a pair of RNLs is linked by a nodal surface. We show that this state can be realized in spinless systems with both $PT$ and nonsymmorphic $S_{2z}T$ symmetries, where the $S_{2z}T$-enforced nodal surface prevents the pair of RNLs from annihilation. Each nodal line carries a nontrivial real Chern number $\nu_R=1$, giving rise to topological hinge Fermi arcs located at a pair of $PT$-related hinges. Guided by this construction, we identify the interpenetrated graphene network (IGN) as a promising material realization. First-principles calculations confirm that a pair of nodal lines traversing the Brillouin zone are linked by a nodal surface and each nodal line carries double nontrivial $Z_2$ charges. The bulk-boundary correspondence of IGN manifests as a pair of hinge Fermi arcs together with drumhead surface states. Our work establishes nodal surfaces as a new linking partner for real nodal lines and provides a roadmap for exploring higher-order real topology in carbon-based and other light-element systems.

Halo nuclei are mostly studied through reactions. In breakup reactions, the projectile is sent upon a target, which induces the dissociation of the loosely-bound halo from the core of the nucleus. In most cases, a spectroscopic factor for the core-halo structure is inferred from experimental data. In this contribution, we present a new analysis of the Coulomb breakup of 11Be on 208Pb at 69A MeV measured at RIKEN. We use an effective multi-channel particle-rotor model of the projectile, which accounts for the excitation of the 10Be core. The calculations are in excellent agreement with the data, independently of the value of the spectroscopic factor. This confirms that breakup cross sections are insensitive to this value and that spectroscopic factors cannot be inferred from such measurements. In the near future, we plan to extend this idea to other reactions.

In platform trials, model-based approaches typically estimate treatment effects conditional on calendar time. However, scientific and regulatory interest often lies in treatment effects defined for a target population spanning multiple enrollment periods, requiring explicit consideration of how effects should be averaged across time. This raises two fundamental challenges. The first is the definition of the appropriate target estimand when combining data across multiple periods. The second is the selection of the estimator to be used. In this work, we examine conditional and marginal estimands in platform trials with time trends, and describe target populations of interest. To address the second challenge, we evaluate model-based, G-computation and augmented inverse probability weighting estimators, comparing their bias and variance. We discuss how the choice of estimand, target population and trial data used for estimation affects estimator performance.

Large forecasting applications often combine statistical, machine-learning, and neural models. These families solve the same problem but differ in fitted state, training procedures, and how they parallelize work. Forecasting software must therefore either hide these differences behind a single estimator interface, or keep the families in separate packages, forcing users to rewrite data preparation and evaluation for every package. We present the Nixtlaverse, an ecosystem of open-source Python libraries for time series forecasting, as a case study of a third design: all libraries share the same long-format panel data and keyed forecast outputs, while every model family keeps its own specialized implementation. We demonstrate this design through three use cases on the public M5 competition data. First, we evaluate statistical, machine-learning, and neural models, and an external engine from a separate ecosystem, in a single rolling-origin evaluation with per-series and hierarchy-weighted metrics. Second, we profile runtime and peak memory from 100 to 30,490 series and locate each family's bottleneck: statistical fitting scales approximately linearly in the number of series, feature construction dominates machine-learning memory, and neural training time is nearly independent of panel size under a fixed training budget. Third, we reconcile the forecasts of multiple engines, including the external one, over all 42,840 series of the M5 hierarchy, with sparse reconciliation where dense implementations exhausted memory. These use cases establish the costs, boundaries, and utility of shared data and output contracts. The Nixtlaverse has seen substantial public distribution, scholarly reuse, and adoption through other forecasting frameworks, and is released under permissive open-source licenses with public datasets, reproducible examples, and verifiable benchmark artifacts.

Long-context reasoning faces two complementary bottlenecks: retaining evidence across long inputs and sustaining computation across many reasoning steps. Existing approaches largely address them separately, with external memory extending access to distant evidence and latent reasoning compressing multi-step computation. We introduce LatentHarness, which unifies memory access and latent reasoning as sequential latent action selection. At each internal step, the model chooses THINK for further computation, RECALL from a fast-weight memory of input evidence and intermediate reasoning states, or EXIT to emit the next token. We train this policy with counterfactual policy distillation, which branches every action for one step and scores its effect on the emitted token. These gains teach the policy when memory is more useful than further reasoning, while gradients through counterfactual recall teach which intermediate states should be retained in memory for future use. Across six general and long-context reasoning benchmarks, LatentHarness at 1.4B improves on the strongest baselines by 2.8% and 10.0% relative, respectively, and runs 5.9x faster than the strongest long-context baseline.

Graph generative models increasingly rely on Graph Transformers (GT) to capture complex dependencies among nodes and edges. While deeper architectures should provide greater expressive capacity and a broader receptive field, their effectiveness can decline with depth: repeated self-attention progressively contracts node representations, impeding information flow and gradient propagation. We analyse this phenomenon from a dynamical systems perspective, focusing on how the denoiser's spectral dynamics affect graph generation. We show that standard GT denoisers become increasingly dissipative as depth grows, leading to vanishing gradients and representation collapse. To isolate the effect of these dynamics, we construct a permutation-equivariant GT with inherently stable, non-dissipative transport. We also introduce a damping mechanism that continuously interpolates between non-dissipative and increasingly contractive regimes, enabling a direct assessment of how dissipation influences generation. Experiments on synthetic and molecular graph generation benchmarks show that the gap between these regimes widens with depth: non-dissipative dynamics preserve representation diversity and gradient flow, sustaining strong generative performance, whereas greater contraction progressively impairs it. These findings identify the denoiser's dynamical regime as a key design factor for deep graph generative models.

Learning to generate machining process plans and toolpaths from B-rep CAD requires coupling discrete operation decisions with continuous tool motion as the workpiece evolves. Correctly predicting an operation sequence does not by itself ensure correct material removal, because each toolpath acts on the stock left by preceding cuts. We formulate this problem around persistent manufacturing objects: object identity determines the target of an operation, while the evolving stock state conditions the generation of its toolpath. Based on this formulation, we propose CNCGEN, a dataset and learning framework for three-axis machining. CNCGEN-Dataset contains approximately 50k geometrically verified synthetic machining flows and 800 held-out real CNC records. Each flow aligns B-rep geometry with object-referenced operations, parameterized toolpaths, intermediate stock states, and verification outcomes, enabling supervision of the correspondence between planning decisions and their geometric effects. CNCGEN generates operations and toolpaths for selected objects step by step, updating a compact machining state to guide subsequent predictions. During training, a learned surrogate verifier provides material-removal feedback that links local predictions to their geometric consequences. Experiments on synthetic and held-out real CNC records show that CNCGEN improves the resulting workpiece geometry and reduces residual material and overcut compared with adapted CNC generation baselines.

The two-dimensional sphere embedded in three-dimensional Euclidean space S2, plays a central role in a variety of scientific and engineering domains, including geophysics, planetary science, geodesy, atmospheric physics, quantum chemistry, cosmology, and virtual reality, among many others. As machine learning increasingly permeates these fields, the demand grows for robust tools that process and model functions on the sphere, while respecting the inherent topological and symmetry properties of the domain. We present torch-harmonics, a comprehensive library that offers efficient, differentiable implementations of advanced signal processing and machine learning (ML) methods for spherical data. These include the spherical harmonic transform (SHT), the spherical analogue of the Fourier transform, vector spherical harmonics, discrete-continuous and spectral convolutions, as well as both global and neighborhood spherical attention mechanisms. Beyond traditional representations, torch-harmonics provides the building blocks for state-of-the-art spherical ML architectures such as spherical transformers in order to enable scalable, rotationally-aware learning and inference in modern scientific and engineering applications.

We study deterministic first-order bilevel optimization under weak lower-level convexity, allowing nonconvex lower-level objectives and without assuming strong convexity, the Polyak-Łojasiewicz condition, or an error-bound property. We consider a $\delta$-relaxed Moreau-gap constraint, with $\delta>0$, for the lower-level stationarity condition and propose an inexact variable-smoothing penalty method (IVSP) for computing its approximate Karush-Kuhn-Tucker (KKT) points. For any fixed relaxation level $\delta$, under a standard extended no-nonzero-abnormal-multiplier constraint qualification (ENNAMCQ), we prove finite stabilization of the adaptive penalty parameter and an overall $\widetilde O(\varepsilon^{-3})$ first-order complexity for computing an $\varepsilon$-KKT point. Notably, we give verifiable sufficient conditions for ENNAMCQ covering convex lower-level objectives without nonconstant affine segments (including the strictly convex case), and a class of nonconvex sample-reweighting models. The positive relaxation avoids the intrinsic constraint-qualification degeneracy of the exact Moreau-gap constraint while achieving an $\mathcal O(\sqrt{\delta})$ lower-level near-stationarity guarantee. Numerical experiments on synthetic and real-world bilevel learning problems illustrate the practical performance of IVSP.

Causal loops escape paradoxes if they generate valid probabilities under arbitrary independent local interventions. We show that a deterministic classical process is logically consistent if and only if it contains no signaling loops: cyclic signaling is forbidden but cyclic causation is not, allowing indefinite causal order. Logical consistency is then found to hold exactly when the list of events of a process is pairwise exclusive and complete, providing the first event-based, intervention-free, non-recursive characterization of consistent causal loops. Our results repair a previously introduced intervention-free criterion that misses global loops and reveal that what makes deterministic classical communications without causal order consistent is a multipartite form of Specker's principle: if any two of several questions can be answered jointly, so can all of them.

Claims of quantum advantage rest on the classical hardness of simulating quantum circuits. Magic, operator scrambling, anticoncentration, and non-Gaussianity for fermionic circuits are standard diagnostics of complex quantum dynamics. For pure states, some of these have been rigorously connected to classical simulability. Whether these diagnostics reliably track the limits of efficient classical simulation under noise remains unclear. Here, we show that in noisy Clifford+$T$ and nearest-neighbour matchgate+SWAP circuits, dynamical diagnostics and classical simulability can separate in both directions. In particular, the diagnostics can remain nontrivial after classical simulation becomes efficient, or become trivial before known efficient classical algorithms apply. We trace this mismatch to the different statistical properties they probe: magic and scrambling depend on fourth-order statistics of the Pauli spectrum, whereas the simulation algorithms depend primarily on second-order moments, which local noise suppresses at different rates. Thus, dynamical diagnostics measured on a noisy quantum device do not by themselves constitute evidence for classical hardness.

We show that every irreducible linear differential equations on the Riemann sphere with regular and/or unramified irregular singularities can be transformed into either the trivial equation or a Fuchsian system of $E_8$-fundamental spectral type by a sequence of invertible transformations consisting of confluences, unfoldings, Laplace transformations, gauge transformations and Möbius transformations. The $E_8$-spectral type is uniquely determined by the original equation. A Fuchsian system of $E_8$-fundamental spectral type has three singular points, 0, 1, and $\infty$. The degrees of the minimal polynomials of the residue matrices at 0 and 1 are two and three, respectively, and the sum of the maximal dimensions of eigenspaces of the three residue matrices is not greater than the size of the matrices. The result follows from the correspondence between spectral types of Fuchsian systems and roots of a star-shaped Kac-Moody root system.

User-generated content has become one of the most-consumed content types. However, capturing spatial audio with consumer hardware is still challenging. Given the widespread success of smart earbuds, binaural audio could be a promising option to capture spatial audio on consumer devices, but its inherent signal characteristic limits its usability as a recording format. In this paper, we propose and define a new task, Binaural to Ambisonics conversion (Bin2Ambi). In our proposed system, we exploit simultaneously captured head-tracking data provided from the motion sensors in smart earbuds. We show that this motion data help resolve the inherent directional uncertainty of two-channel binaural audio due to front-back localization ambiguities and lateral errors in the cone of confusion. Our results show that our system learns directional and diffuse-field information and that head-tracking especially reduces extreme localization errors. Objective metrics and a subjective listening test suggest that the converted Ambisonics soundfield achieves an average directional error of up to $11.8^\circ$ and a perceived spatial quality similar to a DirAC ground-truth model. The proposed algorithm can serve as a baseline for future improvements to this novel Bin2Ambi task.

Supernova remnants (SNRs) are extended radio sources whose morphology and evolution are shaped by the interaction of strong shocks with the surrounding medium. We combined the 0.944 GHz Australian Square Kilometre Array Pathfinder (ASKAP) image from the Evolutionary Map of the Universe (EMU) survey with 0.088-0.200 GHz MWA-GLEAM data to characterize the integrated radio spectrum of G343.1$-$00.7 and perform its first sensitive spatially resolved spectral analysis. We derived an integrated spectral index of $\alpha=-0.50\pm0.01$, confirming the previously reported value with improved accuracy. We redefined the radio morphology of the remnant, providing the first morphological and spectral evidence that filamentary structures overlapping the nearby \hii\ G343.147$-$00.44 are associated with the SNR. Local brightness-brightness analysis revealed flatter-spectrum filaments along the south-eastern shell, consistent with a possible thermal bremsstrahlung contribution from radiative or partially radiative shocks. Comparison with \hi\ and CO data identified gas components kinematically consistent with the SNR distance, although no clear morphological evidence of interaction was found. The gas distribution around the \hii\ suggests an expanding \hi\ shell partially embedded within its parent molecular cloud. Theoretical models place G343.1$-$00.7 between the late Sedov and early pressure-driven phases, suggesting different evolutionary stages across the remnant. These results highlight the potential of high-resolution ASKAP observations for spatially resolved studies of poorly characterized SNRs in complex Galactic environments.

Coherent anti-Stokes Raman scattering (CARS) microscopy enables label-free chemical imaging of biological samples and materials. Conventionally, CARS is implemented using a point-scanning approach that probes a single vibrational mode at a time. Hyperspectral CARS enhances chemical specificity by sequentially addressing multiple Raman modes. However, the measured CARS intensity is distorted by an undesired non-resonant background, which broadens and shifts the Raman peaks, thereby hindering the interpretability of hyperspectral images. Here, we introduce a phase-sensitive, wide-field hyperspectral CARS microscopy scheme that suppresses the non-resonant background. The proposed approach combines three key components: a high-power picosecond tunable optical parametric amplifier (OPA), speckle illumination, and quantitative phase imaging based on quadriwave lateral shearing interferometry (QLSI). The high-power OPA provides the peak power required for efficient nonlinear excitation. Speckle illumination distributes the optical energy over the objective back pupil and mitigates coherent imaging artifacts. QLSI enables the simultaneous measurement of the amplitude and phase of the CARS field, thereby allowing separation of resonant and non-resonant contributions, without the need for an external reference beam. This unique combination enables practical phase-resolved CARS imaging over a field of view exceeding $60 \times 60~\text{\textmu m}^2$ at a frame rate of 1.4 Hz. We illustrate the approach by acquiring hyperspectral images of microplastics and liver steatosis across the entire CH-stretching region.

In the context of a spatially flat FLRW background, we perform a symmetry classification of $k$-essence models with Lagrangian densities of the form $f_1(\phi)R+f_2(\phi,X)$. The kinetic term $X$ is introduced as an independent degree of freedom via a Lagrange multiplier, and the lapse function is treated as a dynamical variable. The symmetry analysis is applied to the constrained system prior to any gauge fixing. This approach reveals symmetries which otherwise are lost when the lapse is fixed at the level of the action. The derived families of $k$-essence models admitting nontrivial symmetries fall to two general classes: minimally coupled and nonminimally coupled theories to gravity. We use the corresponding Noetherian conservation laws to derive exact cosmological solutions. We find that, when the numerical value of the conserved charges is zero, the field equations reduce to an algebraic relation. We subsequently derive power-law expressions for the scale factor with exponents determined by the particular $k$-essence function.

We study a high-dimensional Kuramoto model with attractive pairwise coupling and a repulsive higher-order effect. Although the pairwise interaction favors synchronization, the higher-order interaction prevents complete synchronization and selects an intermediate level of coherence corresponding to the balanced equilibria. For the unit sphere of dimension at least two, we provide an explicit basin of attraction for the balanced equilibria. We also classify nonzero-mean equilibria and show that all non-balanced equilibria are linearly unstable. This indicates that balanced states are the only natural stable candidates in the higher dimensions. In contrast, on the circle, the same model reduces to a higher-order Kuramoto-type model which exhibits a qualitatively different selection mechanism that depends fundamentally on the dimension. In this case, population imbalance at finite $N$ prevents exact stationary balance and instead generates a common angular drift producing phase-locked states. We construct basins of attraction for two-cluster locked states, in which the phases split into two groups, and identify the corresponding finite-$N$ selected values of the order parameters. This phenomenon is called continuous-versus-quantized order-parameter selection. We further demonstrate, through a root-selection mechanism, why such two-cluster locked states are typically observed in most simulations. Finally, we show that two-cluster locked states with a large population imbalance are linearly unstable, which explains why only certain locked states are dynamically robust.

Cooperative language-model agents must coordinate over long horizons and adapt to changing environments and to partners with unfamiliar conventions, yet existing agents map observations to actions without separating persistent coordination strategies from their tactical execution. We introduce OverForge, a training-free hierarchical architecture that separates strategic reasoning over roles and divisions of labour from tactical reasoning over actions within each agent's private, partner-conditioned world model. A metacognitive Prefrontal Cortex Module couples the two levels by forming strategy-action branches, imagining their consequences with a forward model, and committing when confident. In OvercookedV2, OverForge delivers 7 soups in a connected kitchen versus 3 for each flat LLM baseline, retains agreed roles, and adopts roles proposed by unfamiliar partners. Ablations and a fixed-strategy probe show that persistent strategies guide tactical adaptation while each reasoning level contributes to coordination. Memory restarts show that cross-episode partner knowledge supports task performance and partner prediction, linking the hierarchy to continual adaptation.

The known proofs of the gradient estimate for complex Hessian equations rely on the second order estimate of Hou-Ma-Wu and on a Liouville theorem of Dinew-Kolodziej. In this paper, we give a direct proof of the gradient estimate for concave fully nonlinear elliptic equations on compact Hermitian manifolds, which uses neither the Hou-Ma-Wu estimate nor the Liouville theorem. Instead, the proof combines a stability estimate with a comparison argument against nearby smooth admissible functions. We also give an independent proof of the complex Hessian estimate on compact Kähler manifolds, based on Dirichlet Green's functions, under additional assumptions on the operator.

Let $N$ be a square-free, positive integer and let $p\ge 5$ be a prime dividing $N$. In this work, we construct semistable models of Atkin–Lehner quotients of $X_0(N)$ over the Witt ring $W(\overline{\mathbb{F}}_p)$ by taking the corresponding quotients of the Deligne–Rapoport model of $X_0(N)$. We give an explicit description of the special fiber of the models and obtain genus formulae for the Atkin–Lehner quotients of $X_0(N)$.

The analysis of observable phenomena (for instance, in biology or physics) allows the detection of dynamical behaviours. If the conditions are ideal and the number of observations is sufficient, we can represent these phenomena by a dynamical system, also called a functional digraph, that is to say a graph where each node has out-degree exactly one. Up to isomorphism, these dynamical systems, equipped with disjoint union as addition and direct product as multiplication, form a commutative semiring. Several previous studies on this semiring have aimed to establish algebraic properties (primality, injectivity) or complexity results (division, factorisation). However, no work has yet been conducted on graphs derived from imperfect observations that result in missing transitions or nodes, in other words, in cases where each node in the graph has an out-degree of at most one. In this case, we say that the system is partial. In this paper, we show that partial dynamical systems, up to isomorphism and equipped with the same addition and multiplication, still form a commutative semiring. We then characterise the prime elements of this semiring, which differ from those of the semiring of dynamical systems. Finally, we highlight two properties shared by both semirings. First, injective univariate polynomials admit the same characterisation in both. Second, division can be computed in polynomial time for partial dynamical systems if and only if it can be for dynamical systems.

Strong unrestricted adversarial attacks can distort the primary object of an image, hereafter referred to as the subject. To preserve subject integrity without compromising attack magnitude, we introduce the carrier: a secondary visual element that provides an auxiliary region to facilitate the attack under global classifier guidance. We demonstrate three key findings: 1. A carrier mitigates subject distortion by absorbing a larger share of globally normalized attack updates. 2. A carrier improves cross-model transferability, governed by the strength of target-related features that balance semantic separation and transfer performance. 3. Successful targeted attacks retain the personalized subject as the primary content perceived by humans while successfully misleading the classifier. Our results demonstrate that a visually secondary carrier offers an auxiliary spatial pathway for adversarial changes, enabling strong and transferable attacks while improving subject preservation.

Synthetic speech attribution aims to identify the generative system responsible for a speech signal, but current approaches typically rely on black-box neural networks that provide limited insight into their decisions. This work investigates prototype-based networks as an interpretable alternative, where predictions are grounded in comparisons with representative training examples. We adapt ProtoPNet to spectrogram-based speech representations and evaluate the proposed framework on the MLAAD dataset under closed-set, cross-lingual, and open-set conditions. Experiments show that prototype-based reasoning achieves competitive or improved attribution performance compared with the baseline while enabling example-based explanations. These results highlight that interpretability and performance can be jointly achieved in synthetic speech attribution through prototype-based modeling.

We give a new proof of the classification of unipotent representations for split finite groups of Lie type. To do this, we first construct an (infinity, 2)-categorification of Lusztig's map from the Hecke algebra to the asymptotic algebra. Then we use the theory of weights to show that it induces an isomorphism after applying trace of Frobenius.

We briefly review the predictions of holographic QCD in hard-wall AdS/QCD models for the contribution of pseudoscalars and axial-vector mesons to the hadronic light-by-light amplitude and consequently to the muon anomalous magnetic moment, which allows one to satisfy the Melnikov-Vainshtein constraint in a purely hadronic model. Moreover, we discuss the role of tensor mesons in satisfying the remaining short-distance constraints, and provide details for a minimal model of tensor mesons that turns out to agree remarkably well with available data from BELLE as well as preliminary low-$Q^2$ data from BESIII. In an appendix, we also provide analytical results for the short-distance limits in hQCD and compare with the leading-order OPE result as given by the massless quark loop.

Grid-forming (GFM) inverters are gaining attention as a promising alternative for conventional synchronous generators in the modern power systems. Unlike conventional synchronous generators, GFM inverters have limited overcurrent capability that makes them vulnerable during large disturbances. During disturbances i.e., voltage and frequency dips, GFM inverters are pushed into current-limited operation to protect the semiconductor switches. This causes the internal angle of the GFM inverters to accelerate and lose synchronism with the rest of the grid. To address this limitation, this article proposes a transient triggering grid-forming (TTGFM) synchronization control to enhance the synchronization stability performance under voltage and frequency dips. This method uses two feedback signals; terminal voltage and the difference between unsaturated and saturated power to adjust the internal angle which is generated by power synchronization loop (PSL) of the GFM inverters. These two signals manipulates the internal reference angle generation that act as a virtual braking mechanism. This mechanism limits the angle acceleration during voltage and frequency dips without requiring an extra supervisory signal or parameters to tune. The proposed method is benchmarked against two state-of-the-art synchronization enhancement schemes and validated through high-fidelity electromagnetic transient (EMT) simulations with a grid dynamic equivalent (GDE) model in MATLAB/Simulink\textsuperscript{\textregistered}. An analytical framework is developed to derive the synchronization instability mechanism and the critical limits of the stability margins. Generalization of the GDE model further shows that the TTGFM control is not restricted to a single configuration but is applicable to any standard benchmark system.