October 2025 arXiv papers — page 84

Toshihiro Sato, Mengli Hu, Ion Cosma Fulga, Oleg Janson

Altermagnets are a novel class of fully spin-compensated magnetic materials that nevertheless have spin-split electronic bands, offering novel perspectives for spintronics applications. Based on a rigorous analysis of altermagnetic many-body models and their symmetry we establish the important role of two fundamental types of polarizations in altermagnetic i

Motoo Suzuki, Ling-Xiao Xu

Motivated by recent progress in the spurion analysis of non-invertible selection rules (NISRs) arising from near-group fusion algebras, we further generalize the framework to a class of NISRs obtained from $\mathbb{Z}_2$ orbifolding of a $\mathbb{Z}_M$ symmetry, denoted as $\mathbb{Z}_M/\mathbb{Z}_2$. Many structural features are carried over: for instance,

Bruno M. Celiz, Julio F. Navarro, Mario G. Abadi

We use the TNG50 cosmological hydrodynamic simulation to study the accreted stellar component and stellar haloes of isolated galaxies spanning a wide range of masses ($10^8<M_*/M_\odot<10^{11}$). We find that stars formed in the main progenitor (i.e., in-situ stars) typically dominate the inner regions as far as $\sim$10 half-light radii from the centre, imp

Chiara E. Scardoni, Giovanni P. Rosotti, Cathie J. Clarke, Enrico Ragusa

Recent studies on planet-dominated Type II migration demonstrated the presence of a correlation between the direction of planet migration and the parameter K describing the depth of the planetary gap. It was found that high (low) value for K correspond to outward (inward) migration. In this paper we aim at understanding the mechanism driving inward/outward m

Tushar Gupta, Matti Heikinheimo, Katri Huitu, Harri Waltari

We investigate the possibility of saturating the relic density bound with light Higgsinos. When the minimal supersymmetric Standard Model is extended with right-handed neutrino superfields and the seesaw scale is very low, right-handed sneutrinos can be produced via the freeze-in mechanism. In such a case we can have essentially two independent sources for d

Sara Azizi, Swapnil Shankar, Philipp Mösta, Roland Haas

Relativistic macroscopic plasma dynamics can be described by general-relativistic magnetohydrodynamics. In many high-energy astrophysical settings, such as the interior dynamics of magnetized stars, the ideal GRMHD approximation, in which we assume infinite conductivity, provides an excellent description. However, ideal GRMHD neglects resistive effects that

Sayak Guha Roy, Vaibhav Sharma, Kaidi Xu, Umberto Borla

The $\mathbb{Z}_2$ lattice gauge theory is a paradigmatic model that exhibits gauge-field-mediated-confinement of pairs of particles into mesons, drawing connections to quantum chromodynamics. In the absence of any additional attractive interactions between particles, mesons are not known to bind in this model. Here, we show that resonant pair-production ter

Cristhiam Lopez-Arcos

In this work, we compute the one-loop four-graviton amplitudes with a massive abelian vector field (Proca) circulating in the loop. Instead of the conventional Einstein-Hilbert formulation, we employ the Landau-Lifshitz metric density approach, coupling gravity to the Proca field. Within this framework, we found that the resulting contact-terms-free $n$-gon

Lisa Blomberg, Kareem El-Badry, Bethany Ludwig, Maria Drout

Stripped stars are hot, helium-rich stars formed when binary interactions remove a star's hydrogen envelope. While low-mass ($\lesssim 1 M_\odot$) and high-mass ($\gtrsim 8 M_\odot$) stripped stars are well studied as hot subdwarfs and Wolf-Rayet stars, their intermediate-mass counterparts ($1-8 M_\odot$) have only recently been discovered. The Stripped-Star

Martín de los Rios, Serafina Di Gioia, Fabio Iocco, Roberto Trotta

Machine learning has the potential to improve the reconstruction of the dark matter profile of galaxies with respect to traditional methods, like rotation curves. We demonstrate on the simulation suite Illustris-TNG that a steerable equivariant convolutional neural network (CNN) is able to infer the dark matter profiles within and around individual galaxies

Zekun Zhuang, Daniel Shaffer, Jaglul Hasan, Alex Levchenko

Noncentrosymmetric superconductors (NCSs) with Rashba spin-orbit coupling (SOC) and in-plane magnetic fields have emerged as natural platforms for realizing both the bulk superconducting diode effect (SDE) and the Josephson diode effect (JDE) - phenomena characterized by unequal critical currents in opposite directions due to the simultaneous breaking of tim

Juan Herrero-García, Simone Marciano, Juan Racker, Drona Vatsyayan

We present a simple and broadly applicable extension of the Casas-Ibarra parametrisation that captures the structure of all Majorana neutrino mass models. Building directly on the original formulation, our approach naturally accommodates additional degrees of freedom and provides a unified, minimal framework for parametrising the Yukawa sector. It significan

Liam Keenan, Maximilien Péroux

Given a symmetric monoidal $\infty$-category $\mathscr{E}$, compatible with finite colimits, we show that the functor sending a simplicial object in $\mathscr{E}$ to its skeletal filtration is canonically lax symmetric monoidal. This monoidal structure is the analogue of the one induced by the Eilenberg-Zilber homomorphism from the Dold-Kan correspondence. T

Nicholas Earley, Viraj Karambelkar, Mansi Kasliwal, Kishalay De

We present a near-infrared census of stellar large-amplitude variables (LAVs) observed by the Palomar Gattini-IR (PGIR) surveyor from 2019-2021. Over the three-year time period, PGIR performed a brightness-limited survey of the Northern sky (18,000 sq. deg) to J-band AB magnitudes of $\sim 13$ within and $\sim 15$ outside the Galactic plane. From 70 million

Jose A. R. Cembranos, Álvaro Cendal

Dark matter may not be perfectly stable, and its decay could generate distinctive gravitational-wave signatures. In this work, we present model-independent predictions for the stochastic gravitational-wave background arising from the decay of ultralight dark matter into gravitons. Within this framework, we forecast the sensitivity reach of current and forthc

Federico Esposito, Almudena Alonso-Herrero, Santiago García-Burillo, Ismael García-Bernete

The distribution of molecular gas on small scales regulates star formation and the growth of supermassive black holes in galaxy centers, yet the role of active galactic nuclei (AGN) feedback in shaping this distribution remains poorly constrained. We investigate how AGN influence the small-scale structure of molecular gas in galaxy centers, by measuring the

Paolo Arnaudo, Javier Carballo, Benjamin Withers

We show that retarded Green's functions of black hole spacetimes can be expressed as a convergent mode sum everywhere in spacetime. At late times a quasinormal mode sum converges, while at early times a Matsubara (or, Euclidean) mode sum converges. The two regions are separated by a lightcone which scatters from the black hole potential. The Matsubara sum is

Jiaru Li, Christopher E. O'Connor, Frederic A. Rasio

The orbital architectures of compact exoplanet systems record their complicated dynamical histories. Recent research supports the ``breaking-the-chains'' hypothesis, which proposes that compact systems typically form in chains of mean-motion resonances (MMRs) but subsequently break out on a $\sim 100$Myr timescale. We investigate a scenario for breaking the

Sogoud Sherif, Prakash Sharma, Aman Kumar, Hitesh J. Changlani

The fermionic Hubbard model, when combined with the ingredient of frustration, associated with the breaking of particle-hole symmetry, harbors a rich phase diagram. Aspects of theoretical findings associated with the nature of magnetism and metallicity, in a diverse set of parameter regimes, are now being actively investigated in triangular Hubbard cold atom

Merna Abumusabh, Giulio Dujany, Diego Guadagnoli, Axel Iohner

We introduce a model-independent framework to reinterpret Belle II results using only public data, analytically reconstructing the mapping between true and reconstructed kinematic variables within the statistically dominant Inclusive Tagging Analysis. This enables rare-decay measurements to probe light invisible particles -- such as the QCD axion or axion-li

Roman Berens, Lam Hui, Daniel McLoughlin, Riccardo Penco

We present a unified geometric perspective on the symmetries underlying the spin 0, 1 and 2 static perturbations around a Schwarzschild black hole. In all cases, the symmetries are exact, each forming an SO(3,1) group. They can be formulated at the level of the action, provided the appropriate field variables are chosen. For spin 1 and 2, the convenient vari

Joshua N. Benabou, Anson Hook, Claudio Andrea Manzari, Hitoshi Murayama

The absence of a neutron electric dipole moment (EDM) constrains the quantum chromodynamics (QCD) theta angle to be less than one part in ten billion, posing the Strong $CP$ problem. We revisit two classes of proposed solutions. First, we show that when $P$ or $CP$ is realized as a gauged discrete symmetry - as can arise in quantum gravity - the vacuum neces

Subham Dutta Chowdhury, Sean A. Hartnoll, Aditya Hebbar

Quantum mechanical lattice models with local, bounded interactions obey Lieb-Robinson causality. We show that this implies a domain of analyticity of the retarded Green's function $G^R(\omega,{\bf k})$ of local lattice operators as a function of complex frequency $\omega$ and momentum ${\bf k}$, similar to the lightcone analyticity property of relativistic f

Cyrille Marquet, Yu Shi, Bo-Wen Xiao

We develop a unified resummation framework for heavy-meson pair photoproduction that treats soft-gluon radiation in a massive scheme for $|\boldsymbol q| \lesssim m_Q$ and is consistent with the massless limit $m_Q \ll |\boldsymbol q|$, where $\boldsymbol q$ denotes the transverse momentum of the pair and $m_Q$ the quark mass. This framework describes the fu

Robert Penna

We describe a new tunneling solution for the decay of a cosmic string into a burst of gravitational waves. We find the relevant instanton and compute the tunneling rate. Locally, our solution is just an analytic continuation of the Kerr metric (but there is no black hole in our solution). An interesting feature of our result is that there is a conical singul

Gary T. Horowitz, Maciej Kolanowski, Jorge E. Santos

It is known that linearized perturbations of extremal black holes result in growing curvature on the horizon. However, nonlinear perturbations typically do not evolve to extremal black holes and do not have growing curvature at late times. We show that a large class of nonlinear perturbations of an extremal planar anti-de Sitter black hole does have horizon

Meredith Neyer, Aaron Smith, Mark Vogelsberger, Luz Ángela García

We use the THESAN radiation-hydrodynamics simulations to investigate how Lyman-$\alpha$ emitters (LAEs) trace ionized bubble sizes during the Epoch of Reionization. We generate realistic LAE catalogs by combining accurate intrinsic Ly$\alpha$ production and intergalactic transmission with an empirical model for dust absorption and gas outflows. By calibratin

Bruno Valeixo Bento, Miguel Montero

We establish a no-go theorem in the context of string and M-theory flux compactifications on Riemann-Flat manifolds with Casimir energy. Specifically, we show that no dS minimum exists in this setup in dimension $d>3$. The case of dS$_3$ minima is not excluded, but their actual fate can only be ascertained via an explicit construction. We also point out that

Reinaldo Francener, Victor P. Goncalves, Gabriel Rabelo-Soares

The electromagnetic production of a dilepton pair in the muon - ion scattering, usually denoted muon trident process, is investigated considering the feasibility of studying processes induced by muons at LHC using its far-forward detectors. The total and differential cross - sections are estimated taking into account of the Bethe-Heitler and bremsstrahlung c

Haochen Wang, Yuhao Wang, Tao Zhang, Yikang Zhou

While Multimodal Large Language Models (MLLMs) excel at holistic understanding, they struggle in capturing the dense world with complex scenes, requiring fine-grained analysis of intricate details and object inter-relationships. Region-level MLLMs have been a promising step. However, previous attempts are generally optimized to understand given regions in is

Jaewon Kim

We study the instabilities to the conformal critical point of an exactly solvable family of Gross-Neveu models. Using conformal field theory techniques, we construct the zero-temperature phase diagram and identify the superconducting and charge neutral ordered phases that destabilize the critical point. Both instabilities appear only when the fermions are st

Xiaoyu Liu, Chaoyou Fu, Chi Yan, Chu Wu

Current Vision-Language-Action (VLA) models are often constrained by a rigid, static interaction paradigm, which lacks the ability to see, hear, speak, and act concurrently as well as handle real-time user interruptions dynamically. This hinders seamless embodied collaboration, resulting in an inflexible and unresponsive user experience. To address these lim

Woong-Bae G. Zee, S. Lyla Jung, Sanjaya Paudel, Suk-Jin Yoon

Galactic warps are common in disk galaxies. While often attributed to galaxy--galaxy tides, a non-spherical dark matter (DM) halo has also been proposed as a driver of disk warping. We investigate links among warp morphology, satellite distribution, and large-scale structure using the Sloan Digital Sky Survey catalog of warped disks compiled by Zee et al.\ (

Ziang Zhang, Zehan Wang, Guanghao Zhang, Weilong Dai

Reasoning about dynamic spatial relationships is essential, as both observers and objects often move simultaneously. Although vision-language models (VLMs) and visual expertise models excel in 2D tasks and static scenarios, their ability to fully understand dynamic 3D scenarios remains limited. We introduce Dynamic Spatial Intelligence and propose DSI-Bench,

Zhi Zhang, Yixian Shen, Congfeng Cao, Ekaterina Shutova

Existing parameter-efficient fine-tuning (PEFT) methods primarily fall into two categories: addition-based and selective in-situ adaptation. The former, such as LoRA, introduce additional modules to adapt the model to downstream tasks, offering strong memory efficiency. However, their representational capacity is often limited, making them less suitable for

Saurav Goyal, Roman N. Lee, Sven-Olaf Moch, Vaibhav Pathak

We present the first computation of next-to-next-to-leading order (NNLO) pure QED and mixed QCD$\otimes$QED corrections to unpolarized and polarized semi-inclusive deep-inelastic scattering (SIDIS). Building on our previous NNLO QCD results, these corrections are crucial for improving the theoretical precision. The coefficient functions are derived within th

Akshat Gupta, Jay Yeung, Gopala Anumanchipalli, Anna Ivanova

Growing evidence suggests that large language models do not use their depth uniformly, yet we still lack a fine-grained understanding of their layer-wise prediction dynamics. In this paper, we trace the intermediate representations of several open-weight models during inference and reveal a structured and nuanced use of depth. Specifically, we propose a "Gue

Jeffrey Ouyang-Zhang, Pranav Murugan, Daniel J. Diaz, Gianluca Scarpellini

AlphaFold has transformed protein structure prediction, but emerging applications such as virtual ligand screening, proteome-wide folding, and de novo binder design demand predictions at a massive scale, where runtime and memory costs become prohibitive. A major bottleneck lies in the Pairformer backbone of AlphaFold3-style models, which relies on computatio

Claudio Ranucci, Alessandro Carones, Léo Vacher, Nicoletta Krachmalnicoff

The detection of primordial $B$-mode polarisation of the Cosmic Microwave Background (CMB) is a major observational goal in modern Cosmology, offering a potential window into inflationary physics through the measurement of the tensor-to-scalar ratio $r$. However, the presence of Galactic foregrounds poses significant challenges, possibly biasing the $r$ esti

Anarya Ray, Vicky Kalogera

The fourth gravitational wave transient catalog~(GWTC-4) has enabled empirical probes of the theorized pair-instability gap in the higher end of the binary black hole~(BBH) mass-spectrum. In this letter, using flexibly parametrized models, we show that at present there is no evidence of a sharp drop-off in the spectrum of black hole masses near $~40-50M_{\od

Jizhan Fang, Xinle Deng, Haoming Xu, Ziyan Jiang

Despite their remarkable capabilities, Large Language Models (LLMs) struggle to effectively leverage historical interaction information in dynamic and complex environments. Memory systems enable LLMs to move beyond stateless interactions by introducing persistent information storage, retrieval, and utilization mechanisms. However, existing memory systems oft

Malena Sabaté Landman, Silvia Gazzola

This paper introduces a new class of algorithms for solving large-scale linear inverse problems based on new flexible and inexact Golub-Kahan factorizations. The proposed methods iteratively compute regularized solutions by approximating a solution to (re)weighted least squares problems via projection onto adaptively generated subspaces, where the constraint

Jiayu He, Gabriele Fazio, Matteo G. A. Paris

Multiparameter quantum estimation faces a fundamental challenge due to the inherent incompatibility of optimal measurements for different parameters, a direct consequence of quantum non-commutativity. This incompatibility is quantified by the gap between the symmetric logarithmic derivative (SLD) quantum Cramér-Rao bound, which is not always attainable, and

Yanlin Wang, Rongyi Ou, Yanli Wang, Mingwei Liu

Code translation is a crucial task in software development and maintenance. While recent advancements in large language models (LLMs) have improved automated code translation accuracy, these gains often come at the cost of increased inference latency, hindering real-world development workflows that involve human-in-the-loop inspection. To address this trade-

Tsemo Aristide

This paper investigates the foundations of deep learning through insight of geometry, algebra and differential calculus. At is core, artificial intelligence relies on assumption that data and its intrinsic structure can be embedded into vector spaces allowing for analysis through geometric and algebraic methods. We thrace the development of neural networks f

Qianheng Xu

Over 70 million people worldwide experience stuttering, yet most automatic speech systems misinterpret disfluent utterances or fail to transcribe them accurately. Existing methods for stutter correction rely on handcrafted feature extraction or multi-stage automatic speech recognition (ASR) and text-to-speech (TTS) pipelines, which separate transcription fro

Woong-Bae G. Zee, Suk-Jin Yoon

We present a comprehensive investigation into the influence of stellar bars on star formation (SF) in galaxy pairs, using a large sample of low-redshift galaxies ($0.02$\,$<$\,z\,$<$\,$0.08$) from the DESI Legacy Imaging Surveys DR8. Our analysis examines whether bars enhance or suppress SF during pair interactions, and how these outcomes depend on the star-

Pedro Luís Fonseca, Bruno Lima, João Pascoal Faria

Mobile acceptance testing remains a bottleneck in modern software development, particularly for cross-platform mobile development using frameworks like Flutter. While developers increasingly rely on automated testing tools, creating and maintaining acceptance test artifacts still demands significant manual effort. To help tackle this issue, we introduce AToM

Rishav Sen, Jose Paolo Talusan, Abhishek Dubey, Ayan Mukhopadhyay

High-resolution origin-destination (OD) tables are essential for a wide spectrum of transportation applications, from modeling traffic and signal timing optimization to congestion pricing and vehicle routing. However, outside a handful of data rich cities, such data is rarely available. We introduce MOVEOD, an open-source pipeline that synthesizes public dat

Hamed Jelodar, Mohammad Meymani, Samita Bai, Roozbeh Razavi-Far

This paper introduces SBAN (Source code, Binary, Assembly, and Natural Language Description), a large-scale, multi-dimensional dataset designed to advance the pre-training and evaluation of large language models (LLMs) for software code analysis. SBAN comprises more than 3 million samples, including 2.9 million benign and 672,000 malware respectively, each r

Luca Sacchi, Alfonso Palmieri, Vitthal Mishra, Joon-Suh Park

Metasurfaces -- planar arrays of subwavelength nanostructures -- are typically realized with high-index dielectrics, while low-index platforms are often dismissed for their weaker contrast. Here, we identify and experimentally verify regimes where a low-index platform (SiO$_2$) surpasses a high-index counterpart (TiO$_2$). We demonstrate that a low index sup

David Hokken, Dimitris Koukoulopoulos

Let $A = a_0T^m + \sum_{j=1}^{m-1} a_j (T^{m-j}+T^{m+j}) + T^{2m}+1 \in \mathbf{Z}[T]$ be a monic reciprocal polynomial of degree $2m$ sampled randomly by selecting its coefficients $a_0,a_1,\dots,a_{m-1}$ independently according to a given probability measure $\mu$ on $\mathbf{Z}$. For a wide range of measures $\mu$, we prove that $A$ is irreducible with pr

Omer Angel, Shankar Bhamidi, Serte Donderwinkel, Neeladri Maitra

Motivated by questions in social networks, distributed computing and probabilistic combinatorics, the last few years have seen increasing interest in network evolution models where new vertices entering the system need to make decisions based on a partial snapshot of the current state of the network. This paper considers a specific variant of the classical r

Ling Team, Anqi Shen, Baihui Li, Bin Hu

We present Ring-1T, the first open-source, state-of-the-art thinking model with a trillion-scale parameter. It features 1 trillion total parameters and activates approximately 50 billion per token. Training such models at a trillion-parameter scale introduces unprecedented challenges, including train-inference misalignment, inefficiencies in rollout processi

Nathan Mankovich, Kai-Hendrik Cohrs, Homer Durand, Vasileios Sitokonstantinou

Earth observation involves collecting, analyzing, and processing an ever-growing mass of data. This planetary data is crucial for addressing relevant societal, economic, and environmental challenges, ranging from environmental monitoring to urban planning and disaster management. However, its high dimensionality entails significant feature redundancy and com

Yu-Han Yang, Eleonora Troja, Marko Ristić, Muskan Yadav

AT2025ulz is an optical/near-infrared transient discovered during follow-up of the candidate gravitational wave (GW) event S250818k. Its young age ($\lesssim$1 d), rapid decline and strong color evolution over the first 48 hr classify it as a potential kilonova candidate. In this work, we present the results of our observing campaign, carried out with the Gr

Malena Sabaté Landman

Flexible Krylov methods are a common standpoint for inverse problems. In particular, they are used to address the challenges associated with explicit variational regularization when it goes beyond the two-norm, for example involving an $\ell_p$ norm for $0 < p \leq 1$. Moreover, inner-product free Krylov methods have been revisited in the context of ill-pose

Nutkritta Kraipatthanapong, Natthaphat Thathong, Pannita Suksawas, Thanunnut Klunklin

This paper presents a novel Lyapunov-Based Quantum Reinforcement Learning (LQRL) framework that integrates quantum policy optimization with Lyapunov stability analysis for continuous-time vehicle control. The proposed approach combines the representational power of variational quantum circuits (VQCs) with a stability-aware policy gradient mechanism to ensure

Shuofeng Zhang, Ard Louis

In this position paper, we argue that many post-mortem generalization measures -- those computed on trained networks -- are \textbf{fragile}: small training modifications that barely affect the performance of the underlying deep neural network can substantially change a measure's value, trend, or scaling behavior. For example, minor hyperparameter changes, s

Rongyuan Wu, Lingchen Sun, Zhengqiang Zhang, Shihao Wang

Benefiting from pre-trained text-to-image (T2I) diffusion models, real-world image super-resolution (Real-ISR) methods can synthesize rich and realistic details. However, due to the inherent stochasticity of T2I models, different noise inputs often lead to outputs with varying perceptual quality. Although this randomness is sometimes seen as a limitation, it

Kumbha Nagaswetha

High Dynamic Range (HDR) imaging aims to reproduce the wide range of brightness levels present in natural scenes, which the human visual system can perceive but conventional digital cameras often fail to capture due to their limited dynamic range. To address this limitation, we propose a deep learning-based multi-exposure fusion approach for HDR image genera

M. Koshelev

In this paper we obtain the stability theorem for the independence number of $G(n, r, 1)$ graphs. This result was previously stated in the paper of M. Pyaderkin but the proof there was incorrect. We introduce the correct proof of the key lemma and thus finally complete the proof of this theorem.

Chenghao Zhu, Meiling Tao, Tiannan Wang, Dongyi Ding

Faithfully personalizing large language models (LLMs) to align with individual user preferences is a critical but challenging task. While supervised fine-tuning (SFT) quickly reaches a performance plateau, standard reinforcement learning from human feedback (RLHF) also struggles with the nuances of personalization. Scalar-based reward models are prone to rew

Arian Vezvaee, Cesar Benito, Mario Morford-Oberst, Alejandro Bermudez

Demonstrating subthreshold scaling of a surface-code quantum memory on hardware whose native connectivity does not match the code remains a central challenge. We address this on IBM heavy-hex superconducting processors by co-designing the code embedding and control: a depth-minimizing SWAP-based "fold-unfold" embedding that uses bridge ancillas, together wit

Diego Ariel Sotelo Carrillo, Omar Pedraza, L. A. López, R. Arceo

We compute the quasi-normal modes of scalar, electromagnetic, and gravitational perturbations of an Ay\'on-Beato-Garc\'ia black hole surrounded by quintessence using the Asymptotic Iteration Method (30 iterations). The results are compared with the WKB approximation to evaluate the robustness and accuracy of AIM. The results show that the real part of the fr

Ryan Teoh, Sander Tonkens, William Sharpless, Aijia Yang

Hamilton-Jacobi (HJ) Reachability offers a framework for generating safe value functions and policies in the face of adversarial disturbance, but is limited by the curse of dimensionality. Physics-informed deep learning is able to overcome this infeasibility, but itself suffers from slow and inaccurate convergence, primarily due to weak PDE gradients and the

Selina Zhou, Maximilian Shen, Ryoto Sekine, Nicolas Englebert

Supercontinuum light sources are widely used for applications ranging from imaging to sensing and frequency comb stabilization. The most common mechanisms for their generation rely on cubic nonlinearities, for instance in crystals, optical fibers, and integrated photonics. However, quadratic supercontinuum generation (QSCG) offers potential for enhanced ener

Mostapha Kalami Heris

Large Language Models (LLMs) are central to reasoning, writing, and decision-support workflows, yet users lack consistent control over how they reason and express outputs. Conventional prompt engineering relies on verbose natural-language instructions, limiting reproducibility, modularity, and interpretability. This paper introduces Prompt Decorators, a decl

Pawel Morzywolek, Peter B. Gilbert, Alex Luedtke

We provide an inferential framework to assess variable importance for heterogeneous treatment effects. This assessment is especially useful in high-risk domains such as medicine, where decision makers hesitate to rely on black-box treatment recommendation algorithms. The variable importance measures we consider are local in that they may differ across indivi

Dipranjan Chatterjee, Benjamin M. Huddart, Hank C. H. Wu, Dharmalingam Prabhakaran

A spin-polarized muon implanted into a fluoride forms a coupled F--$\mu$--F complex in which the muon spin and neighbouring fluorine nuclear spins become entangled. Here we apply radio-frequency (RF) excitation to this coupled system and use the three-dimensional distribution of emitted positrons to reconstruct the time-dependent evolution of the muon spin p

Jingya Cheng, Alaleh Azhir, Jiazi Tian, Hossein Estiri

Counterfactual inference provides a mathematical framework for reasoning about hypothetical outcomes under alternative interventions, bridging causal reasoning and predictive modeling. We present a counterfactual inference framework for individualized risk estimation and intervention analysis, illustrated through a clinical application to post-acute sequelae

Ling Xing, Rui Yan, Alex Jinpeng Wang, Zechao Li

People see text. Humans read by recognizing words as visual objects, including their shapes, layouts, and patterns, before connecting them to meaning, which enables us to handle typos, distorted fonts, and various scripts effectively. Modern large language models (LLMs), however, rely on subword tokenization, fragmenting text into pieces from a fixed vocabul

Jacob S. Merson, Cameron W. Smith, Mark S. Shephard, Fuad Hasan

This paper presents the Parallel Coupler for Multimodel Simulations (PCMS), a new GPU accelerated generalized coupling framework for coupling simulation codes on leadership class supercomputers. PCMS includes distributed control and field mapping methods for up to five dimensions. For field mapping PCMS can utilize discretization and field information to acc

Yubin Zheng, Pak-Hei Yeung, Jing Xia, Tianjie Ju

Federated learning (FL) enables multiple clients to collaboratively train machine learning models without exposing local data, balancing performance and privacy. However, domain shift and label heterogeneity across clients often hinder the generalization of the aggregated global model. Recently, large-scale vision-language models like CLIP have shown strong

Anthony Conway, Daniel Kasprowski

This paper studies the homotopy and homeomorphism classifications of $4$-manifolds with boundary. Given $4$-manifolds $X_0$ and $X_1$ with fundamental group $π$, we consider the problem of extending a homotopy equivalence $h \colon \partial X_0 \to \partial X_1$ to a homotopy equivalence $X_0 \to X_1$. We solve this problem in broad settings for a class of g

Théophile Chaumont-Frelet, Jérôme Droniou, Simon Lemaire

We establish Maxwell compactness results for the Discrete De Rham (DDR) polytopal complex: sequences in this polytopal complex with bounded discrete $\boldsymbol{H}(\mathbf{curl})$ (resp. discrete $\boldsymbol{H}(\mathrm{div})$) norm and orthogonal to discrete gradients (resp. discrete curls) have $L^2$-relatively compact potential reconstructions. The proof

Jia Zhou, Chang-Xing Ma

In clinical studies with paired organs, binary outcomes often exhibit intra-subject correlation and may include a mixture of unilateral and bilateral observations. Under Donner's constant correlation model, we develop three likelihood-based test statistics (the likelihood ratio, Wald-type, and score tests) for assessing the risk difference between two propor

Hassan Manshouri, Moslem Zarei

We investigate decoherence mechanisms in open quantum systems using quantum field theory techniques and the quantum Boltzmann equation. Specifically, we focus on decoherence through Bremsstrahlung emission, a fundamental process in quantum electrodynamics leading to coherence loss. By applying quantum field theory techniques and quantum Boltzmann equation, w

Aditya Karan, Prabhat Kalle, Nicholas Vincent, Hari Sundaram

Collective action against algorithmic systems provides an opportunity for a small group of individuals to strategically manipulate their data to get specific outcomes, from classification to recommendation models. This effectiveness will invite more growth of this type of coordinated actions, both in the size and the number of distinct collectives. With a sm

Jiawen Zhang, Zhangtao Li, Yuwei Zhang, Hendrik Holz

Dislocations in ceramics have recently gained renewed research interest, in contrast to the traditional belief that ceramics are inherently brittle. Understanding dislocation mechanics in representative oxides is beneficial for effective dislocation engineering. Here, we use MgO single crystals with mechanically seeded dislocation densities from about 10 to

Peter Elbau, Denise Schmutz

This work addresses the problem of uniquely determining a rotational motion from continuous time-dependent measurements within the frameworks of parallel-beam and diffraction tomography. The motivation stems from the challenge of imaging trapped biological samples manipulated and rotated using optical or acoustic tweezers. We analyze the conditions under whi

Yigit Korkmaz, Urvi Bhuwania, Ayush Jain, Erdem Bıyık

Value-based algorithms are a cornerstone of off-policy reinforcement learning due to their simplicity and training stability. However, their use has traditionally been restricted to discrete action spaces, as they rely on estimating Q-values for individual state-action pairs. In continuous action spaces, evaluating the Q-value over the entire action space be

Michael Fraiman, Paulina Hoyos, Tamir Bendory, Joe Kileel

Principal component analysis (PCA) is a fundamental technique for dimensionality reduction and denoising; however, its application to three-dimensional data with arbitrary orientations -- common in structural biology -- presents significant challenges. A naive approach requires augmenting the dataset with many rotated copies of each sample, incurring prohibi

Eric Ramos, Sunny Sun

Given a graph $G$, its independence sequence is the integral sequence $a_1,a_2,...,a_n$, where $a_i$ is the number of independent sets of vertices of size i. In the late 80's Alavi, Erdos, Malde, Schwenk showed that this sequence need not be unimodal for general graphs, but conjectured that it is always unimodal whenever $G$ is a tree. This conjecture was th

Yujie Xing, Xiao Wang, Bin Wu, Hai Huang

Graph Transformers (GTs) have emerged as a powerful paradigm for graph representation learning due to their ability to model diverse node interactions. However, existing GTs often rely on intricate architectural designs tailored to specific interactions, limiting their flexibility. To address this, we propose a unified hierarchical mask framework that reveal

Seunghee Ryu, Donghoon Kwon, Seongjin Choi, Aryan Deshwal

We introduce \textbf{BO4Mob}, a new benchmark framework for high-dimensional Bayesian Optimization (BO), driven by the challenge of origin-destination (OD) travel demand estimation in large urban road networks. Estimating OD travel demand from limited traffic sensor data is a difficult inverse optimization problem, particularly in real-world, large-scale tra

Reilly P. Milburn, Andrew W. Mann, Keighley Rockcliffe, Erin E. Flowers

Gaps in the exoplanet population, such as the Neptunian Desert, point to the importance of mass-loss in sculpting the radii of close-in exoplanets. Young planets ($<$500Myr) offer the opportunity to detect such mass-loss while it is still strong, and to test models of the underlying physical processes. We search for evidence of an H$\alpha$ transit in high-r

Neel Patel, Alexander Wong, Ashkan Ebadi

Tuberculosis remains a critical global health issue, particularly in resource-limited and remote areas. Early detection is vital for treatment, yet the lack of skilled radiologists underscores the need for artificial intelligence (AI)-driven screening tools. Developing reliable AI models is challenging due to the necessity for large, high-quality datasets, w

Jay JH Park, Rebecca K. Metcalfe, Nathaniel Dyrkton, Yichen Yan

Current practices for designing cluster-randomized trials (cRCTs) typically rely on closed-form formulas for power calculations. For cRCTs using covariate-constrained randomization, the utility of conventional calculations might be limited, particularly when data is nested. We compared simulation-based planning of a nested cRCT using covariate-constrained ra

Shuxin Lin, Dhaval Patel, Christodoulos Constantinides

Small Language Models (SLMs) are becoming increasingly popular in specialized fields, such as industrial applications, due to their efficiency, lower computational requirements, and ability to be fine-tuned for domain-specific tasks, enabling accurate and cost-effective solutions. However, performing complex reasoning using SLMs in specialized fields such as

Connor Evans

For a finite Blaschke product $B$ and for an isometry $V$ on an infinite-dimensional separable complex Hilbert space $\mathcal{H}$ we study a sequence $(b_m)_{m=1}^\infty$ of vectors in $\mathcal{H}$, defined by $b_m = B(V^*)e_m$, where $(e_m)_{m=1}^\infty$ is an orthonormal basis in $\mathcal{H}$. We call $(b_m)_{m=1}^\infty$ a Blaschke frame for $B$ with i

Christian Biello, Alessandro Gavardi, Rebecca von Kuk, Matthew A. Lim

We present new state-of-the-art predictions for Standard Model Higgs boson production in association with a bottom-quark pair ($b\bar bH$). Updated cross sections are computed in accordance with the recommendations of the LHC Higgs Working Group, including the use of the PDF4LHC21 set of parton distribution functions, with a center-of-mass energy of 13.6 TeV

Soumyabrata Kundu, Risi Kondor

In contrast to the somewhat abstract, group theoretical approach adopted by many papers, our work provides a new and more intuitive derivation of steerable convolutional neural networks in $d$ dimensions. This derivation is based on geometric arguments and fundamental principles of pattern matching. We offer an intuitive explanation for the appearance of the

Jesús García Fernández, Nasir Ahmad, Marcel van Gerven

Iterative optimization is central to modern artificial intelligence (AI) and provides a crucial framework for understanding adaptive systems. This review provides a unified perspective on this subject, bridging classic theory with neural network training and biological learning. Although gradient-based methods, powered by the efficient but biologically impla

Riccardo Bartocci, Philipp Böer, Tobias Hurth

We report on progress towards including next-to-leading order (NLO) radiative corrections to the subleading-power factorization formula for the $Q_1^{q}-Q_{7\gamma}$ interference contribution in $\bar{B}\to X_s\gamma$ at large photon energies, $m_b - 2E_\gamma = \mathcal{O}(\Lambda_{\rm QCD})$. The novel ingredients for a NLO analysis are the one-loop anomal

Weiqiu You, Siqi Zeng, Yao-Hung Hubert Tsai, Makoto Yamada

Leave-One-Out (LOO) provides an intuitive measure of feature importance but is computationally prohibitive. While Layer-Wise Relevance Propagation (LRP) offers a potentially efficient alternative, its axiomatic soundness in modern Transformers remains largely under-examined. In this work, we first show that the bilinear propagation rules used in recent advan

Tasko Grozdanov, Evgeni Solov'ev

A classical representation for quantum eigenstates of a particle bound in $\lambda z^{2m}$ $(\lambda >0, m=1,2,...)$ potentials is developed. It is represented by ensembles of classical trajectories with energy distributions that can take on negative values, for $m>1$ have integrable singularities at zero energy and whose mean energies coincide with quantum

Marc Gong Bacvanski, Liu Ziyin, Tomaso Poggio

Feedback alignment and related weight-transport-free algorithms are often proposed as biologically plausible alternatives to backpropagation, yet they are typically formulated in discrete phases with implicitly synchronized forward and error signals. We develop a continuous-time model of feedback-alignment-type learning in which neural activities and synapti

Abu Bucker Siddik, Diane Oyen, Soumi De, Greg Olmschenk

Probing properties of neutron stars from photometric observations of these objects helps us answer crucial questions at the forefront of multi-messenger astronomy, such as, what is behavior of highest density matter in extreme environments and what is the procedure of generation and evolution of magnetic fields in these astrophysical environments? However, u

Annapurna Vadaparty, David H. Smith, Samvrit Srinath, Mounika Padala

Generative AI (GenAI) models have broad implications for education in general, impacting the foundations of what we teach and how we assess. This is especially true in computing, where LLMs tuned for coding have demonstrated shockingly good performance on the types of assignments historically used in introductory CS (CS1) courses. As a result, CS1 courses wi