May 2025 arXiv papers — page 117

Pooja Bhattacharjee, Sandra Robles, Stephan A. Meighen-Berger, Francesca Calore

We investigate whether nearby white dwarfs (WDs) can constrain dark matter (DM) interactions with ordinary matter. As experimental sensitivity improves, driven by the Gaia mission, the sample volume of nearby WDs has been increasing over recent years. We carefully select a sample of ten cold, isolated, non-magnetic WDs within 13~pc of the Sun. We model their

Nathaniel Krasner, Nicholas Lanuzo, Antonios Anastasopoulos

Multilingual alignment of sentence representations has mostly required bitexts to bridge the gap between languages. We investigate whether visual information can bridge this gap instead. Image caption datasets are very easy to create without requiring multilingual expertise, so this offers a more efficient alternative for low-resource languages. We find that

Nancy Remage Evans, Alexandre Gallenne, Pierre Kervella, H. Moritz Guenther

The system V350 Sgr has a classical Cepheid for the primary. Interferometry is presented for the system and the full orbit is determined. The mass of the companion has been determined from an {\it IUE} spectrum and comparison with the mass-temperature relation from Detached Eclipsing Binaries. Combined with the mass of the companion (2.6 $\pm$ 0.2 M$_\odot$)

Sean J. Gunderson, Richard Ignace, Walter W. Golay

We present theoretical arguments toward the plausibility of a stellar wind to explain the 16000 km s$^{-1}$ line broadening in the optical spectra of WS 35, the central star in the Pa 30 nebula. The wind model is discussed in the context of super-Eddington flows. We argue that WS 35 potentially occupies a new regime of wind driving theory as the first metal-

Pratik Sathe, Andrew D. King, Susan M. Mniszewski, Carleton Coffrin

Quantum annealing provides a powerful platform for simulating magnetic materials and realizing statistical physics models, presenting a compelling alternative to classical Monte Carlo methods. We demonstrate that quantum annealers can accurately reproduce phase diagrams and simulate critical phenomena without suffering from the critical slowing down that oft

Anirban Biswas, Sougata Ganguly, Dibyendu Nanda, Sujit Kumar Sahoo

Prediction of inflationary observables from the temperature fluctuation of Cosmic Microwave Background (CMB) can play a pivotal role in predicting the reheating dynamics in the early universe. In this work, we highlight how the inflationary observables, in particular the spectral index $n_s$, can play a potential role in constraining the post-inflationary da

Subhasis Maiti, Debaprasad Maity, Rohan Srikanth

In the past two decades, significant advancements have been made in observational techniques to enhance our understanding of the universe and its evolutionary processes. However, our knowledge of the post-inflation reheating phase remains limited due to its small-scale dynamics. Traditional observations, such as those of the Cosmic Microwave Background (CMB)

Zihan Dai, Huanfei Ma

The spiking neural network, known as the third generation neural network, is an important network paradigm. Due to its mode of information propagation that follows biological rationality, the spiking neural network has strong energy efficiency and has advantages in complex high-energy application scenarios. However, unlike the artificial neural network (ANN)

Rohit Goswami

Rigorous performance evaluation is essential for developing robust algorithms for high-throughput computational chemistry. Traditional benchmarking, however, often struggles to account for system-specific variability, making it difficult to form actionable conclusions. We present a Bayesian hierarchical modeling framework that rigorously quantifies performan

Adrian E. Bayer, Francisco Villaescusa-Navarro, Sammy Sharief, Romain Teyssier

We present the first field-level comparison of cosmological N-body simulations, considering various widely used codes: Abacus, CUBEP$^3$M, Enzo, Gadget, Gizmo, PKDGrav, and Ramses. Unlike previous comparisons focused on summary statistics, we conduct a comprehensive field-level analysis: evaluating statistical similarity, quantifying implications for cosmolo

Sarah V. White, Kshitij Thorat, Moses Mogotsi, Rosalind E. Skelton

The GLEAM 4-Jy (G4Jy) Sample is a thorough compilation of the 'brightest' radio sources in the southern sky (Declination < 30 deg), as measured at 151 MHz (S > 4.0 Jy) with the Murchison Widefield Array (MWA), through the GaLactic and Extragalactic All-sky MWA (GLEAM) Survey. In addition to flux-density measurements, the G4Jy catalogue provides host-galaxy i

Matthew Heydeman, Xiaoyi Shi, Gustavo J. Turiaci

We uncover novel features in the spectrum of BPS and near-BPS states in asymptotically $AdS_3 \times S^3 \times S^3 \times S^1$ spacetimes. This follows from a careful analysis of semiclassical and quantum black holes in this theory, which have peculiarities due to the nonlinear large $\mathcal{N}=4$ superconformal symmetry. Notably, we find that the $S^3 \t

Xumin Gao, Mark Stevens, Grzegorz Cielniak

Accurate pest population monitoring and tracking their dynamic changes are crucial for precision agriculture decision-making. A common limitation in existing vision-based automatic pest counting research is that models are typically evaluated on datasets with ground truth but deployed in real-world scenarios without assessing the reliability of counting resu

Christopher Ick, Gordon Wichern, Yoshiki Masuyama, François Germain

The characteristics of a sound field are intrinsically linked to the geometric and spatial properties of the environment surrounding a sound source and a listener. The physics of sound propagation is captured in a time-domain signal known as a room impulse response (RIR). Prior work using neural fields (NFs) has allowed learning spatially-continuous represen

Farshad Rostami Ghadi, Kai-Kit Wong, Masoud Kaveh, F. Javier Lopez-Martinez

This letter introduces the concept of fluid integrated reflecting and emitting surface (FIRES), which constitutes a new paradigm seamlessly integrating the flexibility of fluid-antenna systems (FASs) with the dual functionality of simultaneous transmitting and reflecting reconfigurable intelligent surfaces (STAR-RISs). The potential of the proposed metasurfa

Christopher Zuo, Chenyi Fei, Alexander E. Cohen, Soohwan Kim

Mosquito-borne diseases cause several hundred thousand deaths every year. Deciphering mosquito host-seeking behavior is essential to prevent disease transmission through mosquito capture and surveillance. Despite recent substantial progress, we currently lack a comprehensive quantitative understanding of how visual and other sensory cues guide mosquitoes to

Ke Sun

The high-dimensional parameter space of deep neural networks -- the neuromanifold -- is endowed with a unique metric tensor defined by the Fisher information. Reliable and scalable computation of this metric tensor is valuable for theorists and practitioners. Focusing on neural classifiers, we return to a low-dimensional space of probability distributions, w

Chiara De Leo, Matteo Martinelli, Rocco D'Agostino, Giulia Gianfagna

Several assumptions at the foundation of the standard cosmological model have as a direct consequence a specific relation between cosmological distances, known as the distance duality relation, whose violation would be a smoking gun of deviations from standard cosmology. We explore the role of upcoming gravitational wave observations in investigating possibl

Tim Pearce, David Souto, Douglas Barrett, Benjamin Lok

We explore the integration of multisensory elements in virtual reality reconstructions of historical spaces through a case study of the Virtual Vauxhall Gardens project. While visual and auditory components have become standard in digital heritage experiences, the addition of olfactory stimuli remains underexplored, despite its powerful connection to memory

C. Tortora, G. Tozzi, G. Agapito, F. La Barbera

Relics are massive, compact and quiescent galaxies that assembled the majority of their stars in the early Universe and lived untouched until today, completely missing any subsequent size-growth caused by mergers and interactions. They provide the unique opportunity to put constraints on the first phase of mass assembly in the Universe with the ease of being

Patricia Sorya

We develop and implement an algorithm that computes the full knot Floer complex of knots of thickness one. As an application, by extending this algorithm to certain knots of thickness two, we show that all but finitely many non-integral Dehn surgery slopes are characterizing for most knots with up to 17 crossings.

Minjae Cho

We show that bootstrap methods based on the positivity of probability measures provide a systematic framework for studying both synchronous and asynchronous nonequilibrium stochastic processes on infinite lattices. First, we formulate linear programming problems that use positivity and invariance property of invariant measures to derive rigorous bounds on th

Timo Janßen, Rene Poncelet, Steffen Schumann

We showcase the application of neural importance sampling for the evaluation of NNLO QCD scattering cross sections. We consider Normalizing Flows in the form of discrete Coupling Layers and time continuous flows for the integration of the various cross-section contributions when using the sector-improved residue subtraction scheme. We thereby consider the st

I. Koutsouridou, Á. Skúladóttir, S. Salvadori

Early chemical enrichment processes can be revealed by the careful study of metal-poor stars. In our Local Group, we can obtain spectra of individual stars to measure their precise, but not always accurate, chemical abundances. Unfortunately, stellar abundances are typically estimated under the simplistic assumption of local thermodynamic equilibrium (LTE).

Zhihai Liu, Luyang Wang

Previous investigations have revealed that the Weyl superconductor (WeylSC), realized in a superconductor-topological insulator heterostructure, can exhibit the Landau levels (LLs) of Bogoliubov quasiparticles in the presence of a vortex lattice. Here, we investigate the low-energy quasiparticle (QP) excitations in the mixed state of heterostructure WeylSCs

Cameron R. Morgan, Elizaveta Sazonova, Ian D. Roberts, Michael L. Balogh

We analyze deep ($M_*\gtrsim10^7~{M}_{\odot}$) galaxy stellar mass functions (SMFs) of the Virgo cluster using stellar masses derived as part of the Next Generation Virgo Survey (NGVS). The total SMF has a slope of $\alpha=-1.35^{+0.02}_{-0.02}$ which is similar to or steeper than typical field values. Using deep \ha{} data from the Virgo Environmental Surve

Pranay Gorantla, Abhinav Prem, Nathanan Tantivasadakarn, Dominic J. Williamson

We present a new perspective on the $p$-string condensation procedure for constructing 3+1D fracton phases by implementing this process via the gauging of higher-form symmetries. Specifically, we show that gauging a 1-form symmetry in 3+1D that is generated by Abelian anyons in isotropic stacks of 2+1D topological orders naturally results in a 3+1D $p$-strin

Ioannis Liodakis, Haocheng Zhang, Stella Boula, Riccardo Middei

The origin of the high-energy emission in astrophysical jets from black holes is a highly debated issue. This is particularly true for jets from supermassive black holes that are among the most powerful particle accelerators in the Universe. So far, the addition of new observations and new messengers have only managed to create more questions than answers. H

Matthew M. Murphy, Thomas G. Beatty, Everett Schlawin, Taylor J. Bell

Limb-resolved transmission spectroscopy has the potential to transform our understanding of exoplanetary atmospheres. By separately measuring the transmission spectra of the evening and morning limbs, these atmospheric regions can be individually characterized, shedding light into the global distribution and transport of key atmospheric properties from trans

Andrea Donini, Miguel G. Folgado, Juan Herrero-García, Giacomo Landini

Apart from its gravitational interactions, dark matter (DM) has remained so far elusive in laboratory searches. One possible explanation is that the relevant interactions to explain its relic abundance are mainly gravitational. In this work we consider an extradimensional Randall-Sundrum scenario with a TeV-PeV IR brane, where the Standard Model is located,

Gary T. Horowitz, Donald Marolf, Jorge E. Santos

The Euclidean Einstein-Hilbert action is well-known to be unbounded below and thus to raise many questions regarding the definition of the gravitational path integral. A variety of works since the late 1980's have suggested that this problem disappears when one fixes a foliation of the spacetime and imposes the corresponding gravitational constraints. Howeve

Zachary Llewellyn, Eric Mascot, Oleg A. Tretiakov, Stephan Rachel

Magnetic textures such as skyrmions in thin films grown on substrates possess significant technological potential. Inhomogeneous magnetic structures can be described as homogeneous ferromagnetic order in the presence of anisotropic spin-orbit coupling (SOC). It remains unexplored, however, how this {\it induced} SOC stemming from the magnetic textures intera

Yue Samuel Lu, Dušan Kereš, Philip F. Hopkins, Sam B. Ponnada

Cosmic rays (CRs) are a pivotal non-thermal component of galaxy formation and evolution. However, the intricacies of CR physics, particularly how they propagate in the circumgalactic medium (CGM), remain largely unconstrained. In this work, we study CGM properties in FIRE-2 (Feedback In Realistic Environments) simulations of the same Milky Way (MW)-mass halo

G. Sabatini, E. Bianchi, C. J. Chandler, L. Cacciapuoti

Planet formation around young stars requires the growth of interstellar dust grains from mm-sized particles to km-sized planetesimals. Numerical simulations have shown that large ($\sim$mm-sized) grains found in the inner envelope of young protostars could be lifted from the disc via winds. However we are still lacking unambiguous evidence for large grains i

E. De Rubeis, M. Bondi, A. Botteon, R. J. van Weeren

High sensitivity of modern interferometers is revealing a plethora of filaments surrounding radio galaxies, especially in galaxy cluster environments. The morphology and spectral characteristics of these thin structures require the combination of high-resolution and low frequency observations, which is best obtained using the LOw Frequency ARray (LOFAR) inte

Lucas Kotz, Aurore Courtoy, Pavel Nadolsky, Maximiliano Ponce-Chavez

We present Fanto10, a new ensemble of NLO error parton distribution functions (PDFs) in a charged pion that provides the most detailed estimate of uncertainties from experimental, theoretical, and methodological sources in order to enable faithful comparisons against upcoming precision experiments and ab initio QCD predictions. For the first time, the Fanto1

Jose Maria Sanchez Zaballa, Eugenio Bottacini, Andrea Tramacere

There is general consensus that active galactic nuclei (AGNs) derive their radiating power from a supermassive black hole (SMBH) that accretes matter. Yet, their precise powering mechanisms and the resulting growth of the SMBH are poorly understood, especially for AGNs at high redshift. Blazars are AGNs pointing their jet toward the observer, thus being dete

Christopher P. Herzog, William H. Pannell, Biswajit Sahoo, Andreas Stergiou

Aspects of parity-preserving, charge-conjugation-invariant, three-dimensional conformal field theories (CFTs) with a global $U(1)$ symmetry in the presence of a background magnetic field are investigated. A local effective action is constructed to four-derivative order, based on an assumption that the magnetic field drives the theory into a gapped phase. Thi

Gemma Zhang, Chirag Modi, Oliver H. E. Philcox

Simulation-based inference (SBI) has emerged as a powerful tool for extracting cosmological information from galaxy surveys deep into the non-linear regime. Despite its great promise, its application is limited by the computational cost of running simulations that can describe the increasingly-large cosmological datasets. Recent work proposed a hybrid SBI fr

Paula D. López, Francesca Fragkoudi, Sofía A. Cora, Cecilia Scannapieco

Boxy/peanut (b/p) or X-shaped bulges have been extensively explored with theory and numerical simulations of isolated galaxies. However, it is only recently that advances in hydrodynamical cosmological simulations have made it possible to explore b/p bulges in a cosmological setting, with much remaining to be understood about their formation and evolution. B

Jakob Stegmann, Davide Gerosa, Isobel Romero-Shaw, Giulia Fumagalli

It remains an open question whether the binary black hole mergers observed with gravitational-wave detectors originate from the evolution of isolated massive binary stars or were dynamically driven by perturbations from the environment. Recent evidence for non-zero orbital eccentricity in a handful of events is seen as support for a non-negligible fraction o

Ivana Babić, Fabian Schmidt, Beatriz Tucci

We present results of field-level inference of the baryon acoustic oscillation (BAO) scale $r_s$ on rest-frame dark matter halo catalogs. Our field-level constraint on $r_s$ is obtained by explicitly sampling the initial conditions along with the bias and noise parameters via the LEFTfield EFT-based forward model. Comparing with a standard reconstruction pip

Madelyn Cain, Dolev Bluvstein, Chen Zhao, Shouzhen Gu

Quantum error correction (QEC) is required for large-scale computation, but incurs a significant resource overhead. Recent advances have shown that by jointly decoding logical qubits in algorithms composed of transversal gates, the number of syndrome extraction rounds can be reduced by a factor of the code distance $d$, at the cost of increased classical dec

Pavel Rumiantsev, Mark Coates

Neural Architecture Search (NAS) is a powerful tool for automating architecture design. One-Shot NAS techniques, such as DARTS, have gained substantial popularity due to their combination of search efficiency with simplicity of implementation. By design, One-Shot methods have high GPU memory requirements during the search. To mitigate this issue, we propose

Vinay Samuel, Harshita Diddee, Yiming Zhang, Daphne Ippolito

Aligning language models (LMs) with user intent is becoming increasingly relevant to enhance user experience. This calls for designing methods that can allow users to control the properties of the language that LMs generate, for example, controlling the length of the generation or the complexity of the language that gets chosen. Most existing work attempts t

Zhengyang Geng, Mingyang Deng, Xingjian Bai, J. Zico Kolter

We propose a principled and effective framework for one-step generative modeling. We introduce the notion of average velocity to characterize flow fields, in contrast to instantaneous velocity modeled by Flow Matching methods. A well-defined identity between average and instantaneous velocities is derived and used to guide neural network training. Our method

Dulhan Jayalath, Gilad Landau, Oiwi Parker Jones

Despite major advances in surgical brain-to-text (B2T), i.e. transcribing speech from invasive brain recordings, non-invasive alternatives have yet to surpass even chance on standard metrics. This remains a barrier to building a non-invasive brain-computer interface (BCI) capable of restoring communication in paralysed individuals without surgery. Here, we p

Xiaoyuan Liu, Tian Liang, Zhiwei He, Jiahao Xu

Large Language Models (LLMs) show great promise in complex reasoning, with Reinforcement Learning with Verifiable Rewards (RLVR) being a key enhancement strategy. However, a prevalent issue is ``superficial self-reflection'', where models fail to robustly verify their own outputs. We introduce RISE (Reinforcing Reasoning with Self-Verification), a novel onli

Liyan Tang, Grace Kim, Xinyu Zhao, Thom Lake

Chart understanding presents a unique challenge for large vision-language models (LVLMs), as it requires the integration of sophisticated textual and visual reasoning capabilities. However, current LVLMs exhibit a notable imbalance between these skills, falling short on visual reasoning that is difficult to perform in text. We conduct a case study using a sy

Samuel Goldstein, Oliver H. E. Philcox, Emanuele Fondi, William R. Coulton

Oscillations in the primordial bispectrum are sourced by a range of inflationary phenomena, including features in the inflaton potential and interactions with massive fields through the Cosmological Collider scenario. These signatures offer a powerful window into early-universe physics. In this work, we study how oscillations of the form $\lim_{q\ll k}B(q,k)

Ruoshi Jiang, Zhiyu Fan, Bartomeu Monserrat, Wei Ku

We report an unexpected trans-proximate interlayer correlation (stronger correlation between disjoint layers than the adjacent ones) in the high-pressure phase of the recently discovered La$_4$Ni$_3$O$_{10}$ superconductors. Accompanied by an unusual pressure-induced fractionalization of Ni$^{2+}$ ionic spin from the standard spin-1 to spin-$\frac{1}{2}$, th

Abhay Deshpande, Yuquan Deng, Arijit Ray, Jordi Salvador

We present GrasMolmo, a generalizable open-vocabulary task-oriented grasping (TOG) model. GraspMolmo predicts semantically appropriate, stable grasps conditioned on a natural language instruction and a single RGB-D frame. For instance, given "pour me some tea", GraspMolmo selects a grasp on a teapot handle rather than its body. Unlike prior TOG methods, whic

Ruoyu Wang, Yi Ma, Shenghua Gao

Currently almost all state-of-the-art novel view synthesis and reconstruction models rely on calibrated cameras or additional geometric priors for training. These prerequisites significantly limit their applicability to massive uncalibrated data. To alleviate this requirement and unlock the potential for self-supervised training on large-scale uncalibrated v

Huawei Lin, Tong Geng, Zhaozhuo Xu, Weijie Zhao

Autoregressive (AR) models have recently shown strong performance in image generation, where a critical component is the visual tokenizer (VT) that maps continuous pixel inputs to discrete token sequences. The quality of the VT largely defines the upper bound of AR model performance. However, current discrete VTs fall significantly behind continuous variatio

Penghui Qi, Zichen Liu, Tianyu Pang, Chao Du

Scaling test-time compute is crucial for enhancing the reasoning capabilities of large language models (LLMs). Existing approaches typically employ reinforcement learning (RL) to maximize a verifiable reward obtained at the end of reasoning traces. However, such methods optimize only the final performance under a large and fixed token budget, which hinders e

Dian Shao, Mingfei Shi, Shengda Xu, Haodong Chen

Despite significant advances in video generation, synthesizing physically plausible human actions remains a persistent challenge, particularly in modeling fine-grained semantics and complex temporal dynamics. For instance, generating gymnastics routines such as "switch leap with 0.5 turn" poses substantial difficulties for current methods, often yielding uns

R. James Cotton

Broader access to high-quality movement analysis could greatly benefit movement science and rehabilitation, such as allowing more detailed characterization of movement impairments and responses to interventions, or even enabling early detection of new neurological conditions or fall risk. While emerging technologies are making it easier to capture kinematics

Priyankar Banerjee, Adam Burgess, Julian Wiercinski, Moritz Cygorek

We calculate experimentally measurable signatures of quantum correlations in a coupled molecular dimer that strongly interacts with its vibrational environment. We investigate intensity and mode-resolved photon coincidences for different relative orientations of such dimers, and observe spatio-temporal correlations for various configurations. We find that pr

Mateusz Bystroński, Mikołaj Hołysz, Grzegorz Piotrowski, Nitesh V. Chawla

Data scarcity and class imbalance are persistent challenges in training robust NLP models, especially in specialized domains or low-resource settings. We propose a novel technique, SMOTExT, that adapts the idea of Synthetic Minority Over-sampling (SMOTE) to textual data. Our method generates new synthetic examples by interpolating between BERT-based embeddin

Matthias Göbel, Alejandro Kievsky

Correlation functions as they can be observed in heavy-ion collisions using the femtoscopy technique are a powerful tool to study the interaction among different baryons or mesons. Specifically, the multi-nucleon correlation functions have been under intense experimental and theoretical investigation in the recent years. Due to the interest of using this obs

Meshi Bashari, Roy Maor Lotan, Yonghoon Lee, Edgar Dobriban

Conformal prediction is a framework for predictive inference with a distribution-free, finite-sample guarantee. However, it tends to provide uninformative prediction sets when calibration data are scarce. This paper introduces Synthetic-powered predictive inference (SPI), a novel framework that incorporates synthetic data -- e.g., from a generative model --

Xinzhu Liang, Joseph M. Lukens, Sanjaya Lohani, Brian T. Kirby

This work introduces a new method designed for Bayesian deep learning called scalable Bayesian Monte Carlo (SBMC). The method is comprised of a model and an algorithm. The model interpolates between a point estimator and the posterior. The algorithm is a parallel implementation of sequential Monte Carlo sampler (SMC$_\parallel$) or Markov chain Monte Carlo (

Dian Wang, Boce Hu, Shuran Song, Robin Walters

Recently, equivariant neural networks for policy learning have shown promising improvements in sample efficiency and generalization, however, their wide adoption faces substantial barriers due to implementation complexity. Equivariant architectures typically require specialized mathematical formulations and custom network design, posing significant challenge

Sifeng Shang, Jiayi Zhou, Chenyu Lin, Minxian Li

As the size of large language models grows exponentially, GPU memory has become a bottleneck for adapting these models to downstream tasks. In this paper, we aim to push the limits of memory-efficient training by minimizing memory usage on model weights, gradients, and optimizer states, within a unified framework. Our idea is to eliminate both gradients and

Cristobal Eyzaguirre, Igor Vasiljevic, Achal Dave, Jiajun Wu

We propose a data-driven approach to analyzing query complexity in Video Question Answering (VideoQA). Previous efforts in benchmark design have relied on human expertise to design challenging questions, yet we experimentally show that humans struggle to predict which questions are difficult for machine learning models. Our automatic approach leverages recen

Peace Ayegba, Sofiat Olaosebikan

We study the Student Project Allocation problem with lecturer preferences over Students (SPA-S), an extension of the well-known Stable Marriage and Hospital Residents problem. In this model, students have preferences over projects, each project is offered by a single lecturer, and lecturers have preferences over students. The goal is to compute a stable matc

Lingxiao Du, Fanqing Meng, Zongkai Liu, Zhixiang Zhou

While Multimodal Large Language Models (MLLMs) have achieved impressive progress in vision-language understanding, they still struggle with complex multi-step reasoning, often producing logically inconsistent or partially correct solutions. A key limitation lies in the lack of fine-grained supervision over intermediate reasoning steps. To address this, we pr

Liang Chen, Hongcheng Gao, Tianyu Liu, Zhiqi Huang

Vision-Language Models (VLMs) excel in many direct multimodal tasks but struggle to translate this prowess into effective decision-making within interactive, visually rich environments like games. This ``knowing-doing'' gap significantly limits their potential as autonomous agents, as leading VLMs often performing badly in simple games. To address this, we i

Zhi-Hao Tan, Zi-Chen Zhao, Hao-Yu Shi, Xin-Yu Zhang

The learnware paradigm offers a novel approach to machine learning by enabling users to reuse a set of well-trained models for tasks beyond the models' original purposes. It eliminates the need to build models from scratch, instead relying on specifications (representations of a model's capabilities) to identify and leverage the most suitable models for new

Jonathan E. Moussa

The development of semiempirical models to simplify quantum mechanical descriptions of atomistic systems is a practice that started soon after the discovery of quantum mechanics and continues to the present day. There are now many methods for atomistic simulation with many software implementations and many users, on a scale large enough to be considered as a

Ivan P-Castro, Héctor Aguilera-Trujillo, Hugo García-Compeán, Abdel Pérez-Lorenzana

In this letter, after briefly discussing the Hawking radiation (HR) as a quantum tunneling, we emphasize that special care must be taken when computing temperature corrections in the presence of Lorentz invariance violation (LIV), particularly by ensuring consistency between the modified dispersion relation and the gamma matrix structure used to describe fer

Connor Lennon, Edward Rubin, Glen Waddell

Machine learning (ML) primarily evolved to solve "prediction problems." The first stage of two-stage least squares (2SLS) is a prediction problem, suggesting potential gains from ML first-stage assistance. However, little guidance exists on when ML helps 2SLS$\unicode{x2014}$or when it hurts. We investigate the implications of inserting ML into 2SLS, decompo

Si-Yang Liu, Qile Zhou, Han-Jia Ye

Tabular data, a fundamental data format in machine learning, is predominantly utilized in competitions and real-world applications. The performance of tabular models--such as gradient boosted decision trees and neural networks--can vary significantly across datasets due to differences in feature distributions and task characteristics. Achieving top performan

Samuel Sánchez López, José Jaime Terente Díaz

Primordial scalar perturbations that reenter the horizon after inflation may induce a second-order Gravitational Wave spectrum with information about the primordial Universe on scales inaccessible to Cosmic Microwave Background experiments. In this work, we develop a general framework for the study of Scalar-Induced Gravitational Waves in Palatini $f(R)$ gra

Zhuozhao Hu, Kaishen Yuan, Xin Liu, Zitong Yu

Facial Emotion Analysis (FEA) plays a crucial role in visual affective computing, aiming to infer a person's emotional state based on facial data. Scientifically, facial expressions (FEs) result from the coordinated movement of facial muscles, which can be decomposed into specific action units (AUs) that provide detailed emotional insights. However, traditio

Lotem Peled-Cohen, Maya Zadok, Nitay Calderon, Hila Gonen

Cognitive decline often surfaces in language years before diagnosis. It is frequently non-experts, such as those closest to the patient, who first sense a change and raise concern. As LLMs become integrated into daily communication and used over prolonged periods, it may even be an LLM that notices something is off. But what exactly do they notice--and shoul

Jiajie Zhang, Nianyi Lin, Lei Hou, Ling Feng

Recently, large reasoning models have achieved impressive performance on various tasks by employing human-like deep thinking. However, the lengthy thinking process substantially increases inference overhead, making efficiency a critical bottleneck. In this work, we first demonstrate that NoThinking, which prompts the reasoning model to skip thinking and dire

Artem Riabinin, Egor Shulgin, Kaja Gruntkowska, Peter Richtárik

Recent developments in deep learning optimization have brought about radically new algorithms based on the Linear Minimization Oracle (LMO) framework, such as $\sf Muon$ and $\sf Scion$. After over a decade of $\sf Adam$'s dominance, these LMO-based methods are emerging as viable replacements, offering several practical advantages such as improved memory eff

Yaqian Chen, Hanxue Gu, Haoyu Dong, Qihang Li

Accurately registering breast MR images from different time points enables the alignment of anatomical structures and tracking of tumor progression, supporting more effective breast cancer detection, diagnosis, and treatment planning. However, the complexity of dense tissue and its highly non-rigid nature pose challenges for conventional registration methods

Dongyi Wang, Yuanwei Jiang, Zhenyi Zhang, Xiang Gu

Learning the underlying dynamics of single cells from snapshot data has gained increasing attention in scientific and machine learning research. The destructive measurement technique and cell proliferation/death result in unpaired and unbalanced data between snapshots, making the learning of the underlying dynamics challenging. In this paper, we propose join

Thomas Brüstle, Steve Oudot, Luis Scoccola, Hugh Thomas

Given a finite dimensional, bigraded module over the polynomial ring in two variables, we define its two-parameter count, a natural number, and its end-curves, a set of plane curves. These are two-dimensional analogues of the notions of bar-count and endpoints of singly-graded modules over the polynomial ring in one variable, from persistence theory. We show

Matthias Beck, Emily Clader

The mathematics of musical intervals and scales has been extensively studied. Vastly simplified, our ears seem to prefer intervals whose frequency ratios have small numerator and denominator, such as 2:1 (octave), 3:2 (perfect fifth), 4:3 (perfect fourth), and so on. While there also have been numerous studies on the mathematics of musical chords, very few o

Thangarajah Akilan, Nusrat Jahan, Wandong Zhang

Supervised learning demands large amounts of precisely annotated data to achieve promising results. Such data curation is labor-intensive and imposes significant overhead regarding time and costs. Self-supervised learning (SSL) partially overcomes these limitations by exploiting vast amounts of unlabeled data and creating surrogate (pretext or proxy) tasks t

Saul Kato

We describe a connectionist model that attempts to capture a notion of experience-based problem solving or task learning, whereby solutions to newly encountered problems are composed from remembered solutions to prior problems. We apply this model to the computational problem of \emph{efficient sequence generation}, a problem for which there is no obvious gr

Jinhe Bi, Danqi Yan, Yifan Wang, Wenke Huang

Recent Large Reasoning Models significantly improve the reasoning ability of Large Language Models by learning to reason, exhibiting the promising performance in solving complex tasks. LRMs solve tasks that require complex reasoning by explicitly generating reasoning trajectories together with answers. Nevertheless, judging the quality of such an output answ

R. Luque, C. Piaulet-Ghorayeb, M. Radica, Q. Xue

Recent JWST observations of the temperate sub-Neptune K2-18 b have been interpreted as suggestive of a liquid water ocean with possible biological activity. Signatures of DMS and DMDS have been claimed in the near-infrared (using the NIRISS and NIRSpec instruments) and mid-infrared (using MIRI). However, the statistical significance of the atmospheric imprin

Gabriel Maliakal, Ismail Alkhouri, Alvaro Velasquez, Adam M Alessio

The Maximum Cut (MaxCut) problem is NP-Complete, and obtaining its optimal solution is NP-hard in the worst case. As a result, heuristic-based algorithms are commonly used, though their design often requires significant domain expertise. More recently, learning-based methods trained on large (un)labeled datasets have been proposed; however, these approaches

Nithin Rao Koluguri, Monica Sekoyan, George Zelenfroynd, Sasha Meister

Multi-task and multilingual approaches benefit large models, yet speech processing for low-resource languages remains underexplored due to data scarcity. To address this, we present Granary, a large-scale collection of speech datasets for recognition and translation across 25 European languages. This is the first open-source effort at this scale for both tra

Sebastian Cammerer, Guillermo Marcus, Tobias Zirr, Fayçal Aït Aoudia

We introduce the NVIDIA Sionna Research Kit, a GPU-accelerated research platform for developing and testing AI/ML algorithms in 5G NR cellular networks. Powered by the NVIDIA Jetson AGX Orin, the platform leverages accelerated computing to deliver high throughput and real-time signal processing, while offering the flexibility of a software-defined stack. Bui

Renjie Pi, Felix Bai, Qibin Chen, Simon Wang

The paradigm of using Large Language Models (LLMs) and Multimodal Large Language Models (MLLMs) as evaluative judges has emerged as an effective approach in RLHF and inference-time scaling. In this work, we propose Multimodal Reasoner as a Judge (MR. Judge), a paradigm for empowering general-purpose MLLMs judges with strong reasoning capabilities. Instead of

W. H. Lippincott, H. N. Nelson, D. S. Akerib, C. Amarasinghe

Experimental efforts searching for dark matter particles over the last few decades have ruled out many candidates led by the new generation of tonne-scale liquid xenon. For light dark matter, hydrogen could be a better target than xenon as it would offer a better kinematic match to the low mass particles. This article describes the HydroX concept, an idea to

Xin H. H. Zhang, Daniel Malz, Peter Rabl

We study the collective decay of an initially inverted ensemble of two-level emitters in two distinct scenarios: when coupled to a squeezed photonic reservoir and when interacting with a one-dimensional waveguide. Using a quantum-state diffusion approach to unravel the emission process, we investigate entanglement and classical correlations along individual

Ali Essam Ghareeb, Benjamin Chang, Ludovico Mitchener, Angela Yiu

Scientific discovery is driven by the iterative process of background research, hypothesis generation, experimentation, and data analysis. Despite recent advancements in applying artificial intelligence to scientific discovery, no system has yet automated all of these stages in a single workflow. Here, we introduce Robin, the first multi-agent system capable

C. Murray, C. Combet, C. Payerne, M. Ricci

We present a new method for estimating galaxy cluster masses using weak-lensing magnification. The effect of weak-lensing magnification introduces a correlation between the position of foreground galaxy clusters and the density of background sources. Therefore, cluster masses can be inferred through observations of these correlations. In this work, we introd

Matan Abudy, Orr Well, Emmanuel Chemla, Roni Katzir

State-of-the-art neural networks can be trained to become remarkable solutions to many problems. But while these architectures can express symbolic, perfect solutions, trained models often arrive at approximations instead. We show that the choice of regularization method plays a crucial role: when trained on formal languages with standard regularization ($L_

Benoit Dherin, Michael Munn, Hanna Mazzawi, Michael Wunder

Modern deep learning algorithms use variations of gradient descent as their main learning methods. Gradient descent can be understood as the simplest Ordinary Differential Equation (ODE) solver; namely, the Euler method applied to the gradient flow differential equation. Since Euler, many ODE solvers have been devised that follow the gradient flow equation m

Ewan Davies, Juspreet Singh Sandhu, Brian Tan

We extend the study of the occupancy fraction of the hard-core model in two novel directions. One direction gives a tight lower bound in terms of individual vertex degrees, extending work of Sah, Sawhney, Stoner and Zhao which bounds the partition function. The other bounds the variance of the size of an independent set drawn from the model, which is strictl

Silvia Ferrario Ravasio

Parton showers lie at the core of Shower Monte Carlo event generators, the default theoretical tools used to interpret collider data. In these proceedings, we summarise the strategy of the PanScales collaboration that led to the attainment of the first demonstrably next-to-next-to-leading-logarithmic accurate parton showers.

Shreya Verma, Abhishek Mitra, Qiaohong Wang, Ruhee D'Cunha

One of the primary challenges in quantum chemistry is the accurate modeling of strong electron correlation. While multireference methods effectively capture such correlation, their steep scaling with system size prohibits their application to large molecules and extended materials. Quantum embedding offers a promising solution by partitioning complex systems

Christopher K. Frantz

This article provides an overview of IG Parser, a software that facilitates qualitative content analysis of formal (e.g., legal) rules or informal (e.g., social) norms, and strategies (such as conventions) -- referred to as institutions -- that govern social systems and operate configurally to describe institutional systems. To this end, the IG Parser employ