Latest papers
Time series classification is central to domains like medical signal analysis, industrial monitoring, and sensor-based activity recognition, where class information manifests as localized shapes, specific frequencies, temporal shifts, or…
We prove a desingularization theorem for analytic vortex sheets of the 3D incompressible Euler equations. Starting from an analytic solution of the corresponding Birkhoff-Rott system, we construct, for every sufficiently small thickness…
Hierarchical Reinforcement Learning (HRL) intends to separate strategic planning from primitive execution. It has been widely successful in solving long-horizon and complex tasks, where flat-RL algorithms have difficulty in learning.…
Prior work on human label variation (HLV) in natural language inference (NLI) has often relied on re-annotation resources that select items by disagreement level. An earlier study (arXiv:2607.15870) found that hypotheses containing…
In incidence geometry, a triality is a symmetry cyclically exchanging triples of types of elements. Requiring geometries with trialities to satisfy standard regularity conditions makes their construction highly non trivial, and known…
We show that, for almost-minimizers of the Griffith energy in the plane, the complement of the crack is locally covered by a finite number of John domains where the displacement satisfies H\"older-type estimates. As a consequence, we derive…
Real-world spatial intelligence requires agents to understand scenes from continuous video streams, where objects move, persist, disappear, and reappear over time. While recent spatial foundation models have enabled generalizable…
Simulating active biological tissues, such as the myocardium, requires constitutive models that are both physically faithful and computationally efficient. The most common approach relies on finite element methods that accurately discretize…
Reinforcement learning with verifiable rewards (RLVR) has been established as a viable paradigm for the post-training of Large Language Models (LLMs), including downstream tasks, such as Neural Machine Translation (NMT). With the latest…
After almost a century of speculation, the physical processes responsible for heating the Sun's corona remain uncertain. Observations of coronal loops at extreme ultraviolet and X-ray wavelengths have provided substantial insights, such as…
The magnetism in NaV2O5 results from doping of the narrow split-off conduction band of V2O5, which becomes half-filled and leads to a Mott-insulating splitting of spin resolved bands with anti-ferromagnetic order of the spins along the…
Speculative decoding, in which a lightweight draft model first generates a draft sequence that is then verified in parallel by the target model, has become a prevalent paradigm for accelerating large language model inference. Recent work…
Context: Accretion of He-rich matter onto a low mass carbon-oxygen white dwarf in a binary system with a He-rich donor may lead to an explosive event of Supernova Ia proportion, though with low peak luminosity and peculiar nucleosynthesis.…
We investigate the statistical properties of random quadrangular surfaces generated by different randomization procedures: the Gruzberg-Kl\"umper-Nuding-Sedrakyan (GKNS) construction, and two newly introduced generalizations of dynamical…
We propose an extension of the quarkyonic matter framework that includes $u$, $d$, and $s$ quarks and the full baryon octet. Within this extended framework, we impose beta-equilibrium between baryons and leptons, while determining the quark…
Large language models (LLMs) have been widely applied to automated essay scoring (AES) and automated feedback generation (AFG). However, existing studies rely primarily on prompt engineering or supervised fine-tuning, while systematic…
Heavy-tailed diffusion models replace Gaussian noise by a Gaussian variance mixture: denoising Levy probabilistic models (DLPM) take the mixing variables i.i.d. across coordinates, while Student-t EDM shares one mixing variable per sample.…
The La$_2$NiO$_4$ compound stabilizes in nature in the so-called T structure with octahedral oxygen coordination of the Ni site. Motivated by a similarly existing polymorph for La$_2$CuO$_4$ in the cuprate system, we here study…
In this work, we study a single-source single-server continuous-time status update system where the updates arrive according to a Poisson process and update service times are generally distributed. In our proposed setting, a preemption…
As large language model (LLM) agents move from isolated prompting to longhorizon workflows, failures increasingly arise at the role-to-instance binding boundary, where task-specific role requests must be assigned to concrete agent instances…