arXiv preprints from January 1, 2026 through September 23, 2026 — 02:06:19 EST
Self-supervised pre-training with contrastive learning is a powerful method for learning from sparsely labeled data. However, performance can drop considerably when there is a shift in the distribution of data from training to test time. We study this phenomenon in a setting in which the training data come from multiple domains, and...
We show that the composition of omega-series by surreal numbers, or more generally by elements of any confluent field of transseries, is monotonic in its second argument. In particular, omega-series and LE-series interpreted as functions have the intermediate value property. We also deduce a Taylor approximation theorem for...
The metric MinDist, introduced recently to quantify the distance of an arbitrary Rummy hand from a valid declaration, plays a central role in algorithmic hand evaluation and optimal play. Existing results show that the MinDist of any $13$-card Rummy hand from a single deck is bounded above by $9$. In this paper, we sharpen this bound...
We present 121 days of multi-band (\Bband, \gband, \rband, \iband) optical photometry of the Type Ia supernova SN 2025bvm, obtained with the COLIBRI telescope at OAN-SPM. The light curves show a photometric decline of $Δm_{15}(B) = 0.867 \pm 0.051$~mag, characteristic of a slow-declining Type Ia supernova. After correcting for host...
This work develops a unified nonlinear estimation-guidance-control framework for cooperative simultaneous interception of a stationary target under a heterogeneous sensing topology, where sensing capabilities are non-uniform across interceptors. Specifically, only a subset of agents is instrumented with onboard seekers...
In this paper, we continue the analysis of the effects of semiclassical sub principal controlled quasimodes, approximate solutions to P(h)u(h,b), depending on the subprincipal symbol b, which can give spectral insta bility (pseudospectrum). We consider a pseudodifferential operator, which has double zeros for the principal symbol, p....
Many materials properties depend on higher-order derivatives of the potential energy surface, yet machine learned interatomic potentials (MLIPs) trained with a standard loss on energy, force, and stress errors can exhibit error in curvature, degrading the prediction of vibrational properties. We introduce phonon fine-tuning (PFT),...
Understanding how microorganisms navigate in complex environments is a central question in active matter and biological physics. Phototaxis - the ability to use light as a navigation cue - is a widespread strategy in motile microalgae to optimise photosynthesis and avoid light-induced stress. The microalga Chlamydomonas reinhardtii is...
The celebrated formula of Otter \emph{[Ann. of Math. (2) 49 (1948), 583--599]} asserts that the complete graph contains exponentially many non-isomorphic spanning trees. In this paper, we show that every connected almost regular graph with sufficiently large degree already contains exponentially many non-isomorphic spanning trees....
Accurate modeling of gravitational-wave signals is essential for reliable inference of compact-binary source parameters, particularly for future space-based detectors operating in the milli- and deci-Hertz bands. In this work, we systematically investigate the parameter-estimation biases induced by neglecting orbital eccentricity when...
In this paper, in the first instance the attitude dynamics of the Earth is modelled based on physical principles so as to correctly predict the Chandler wobble and its features such as its period. To this end not only the steady state and dynamic gravity gradient torques were included, but also the effects of centrifugal acceleration...
Multimodal Large Language Models (MLLMs) have demonstrated remarkable performance in mainstream visual understanding tasks, but their ability to process action scenes that contradict everyday common sense remains undertested. To address this gap, we introduce CAIT, a benchmark comprising 400 high-fidelity synthetic scenes focused on...
We investigate the motion of charged particles evolving around a magnetized Kiselev black hole, in the weak magnetic field approximation. The effective potential allows us to study the bound motion and the stable circular orbits. We analyze the impact of combined quintessence and magnetic fields on the epicyclic frequencies. Finally,...
Urban traffic regulation policies are increasingly used to address congestion, emissions, and accessibility in cities, yet their impacts are difficult to assess due to the socio-technical complexity of urban mobility systems. Recent advances in data availability and computational power enable new forms of model-driven,...
The so-called $I$-Love-$Q$ relations link some normalized versions of the moment of inertia, the Love number, and the quadrupole moment of a star. These relations, in principle, enable the inference of two of the quantities given the third. However, their use has been limited because the normalized versions of the multipole moments...
We present a physics-informed Wasserstein GAN with gradient penalty (WGAN-GP) for solving the inverse Chafee--Infante problem on two-dimensional domains with Dirichlet boundary conditions. The objective is to reconstruct an unknown initial condition from a near-equilibrium state obtained after 100 explicit forward Euler iterations of...
This letter presents an analysis of the infrared ($\sim 3-28\,μm$) spectra extracted from the nuclear ($r < 150$ pc) regions of four low-luminosity active galactic nuclei (AGN), observed by JWST NIRSpec/IFU and MIRI/MRS as an extension of the Galaxy Activity, Torus, and Outflow Survey (GATOS). We find that, compared to...
This paper identifies the Stein-exponent of distributed detection when the sensor communicates to the decision center over a discrete memoryless channel (DMC) subject to one of three stringent communication constraints: 1) The number of channel uses of the DMC grows sublinearly in the number of source observations n; 2) The number of...
While NLP typically treats documents as independent and unordered samples, in longitudinal studies, this assumption rarely holds: documents are nested within authors and ordered in time, forming person-indexed, time-ordered $\textit{behavioral sequences}$. Here, we demonstrate the need for and propose a longitudinal modeling and...
We introduce a general method to merge multidimensional equations of state (EoSs) by combining them in a two-fluid equilibrium statistical mixture in the grand canonical ensemble. The merged grand potential density $ω$ is built directly from the input EoSs and the fluid fractions are fixed by minimizing $ω$ at fixed temperature $T$...
We introduce VULCA-Bench, a multicultural art-critique benchmark for evaluating Vision-Language Models' (VLMs) cultural understanding beyond surface-level visual perception. Existing VLM benchmarks predominantly measure L1-L2 capabilities (object recognition, scene description, and factual question answering) while under-evaluate...
The Habitable Worlds Observatory (HWO) is the first NASA Astrophysics flagship mission with a key science goal of searching for signs of life on rocky habitable exoplanets beyond our solar system. The Living Worlds Community Working Group was charged with investigating how HWO could characterize planets orbiting stars in the solar...
The rapid proliferation of misinformation across online platforms underscores the urgent need for robust, up-to-date, explainable, and multilingual fact-checking resources. However, existing datasets are limited in scope, often lacking multimodal evidence, structured annotations, and detailed links between claims, evidence, and...
The computational burden of attention in long-context language models has motivated two largely independent lines of work: sparse attention mechanisms that reduce complexity by attending to selected tokens, and gated attention variants that improve training sta-bility while mitigating the attention sink phenomenon. We observe that...
Vision-Language Models (VLMs) excel at visual description yet remain under-validated for cultural interpretation. Existing benchmarks assess perception without interpretation, and common evaluation proxies, such as automated metrics and LLM-judge averaging, are unreliable for culturally sensitive generative tasks. We address this...