March 2025 arXiv papers — page 165

Ziang Yin, Aashish Poonia, Ashish Reddy Bommana, Xinyu Zhao

The future of artificial intelligence (AI) acceleration demands a paradigm shift beyond the limitations of purely electronic or photonic architectures. Photonic analog computing delivers unmatched speed and parallelism but struggles with data movement, robustness, and precision, while electronic processing-in-memory (PIM) enables energy-efficient computing b

Nicklas Lind, Nilan Paramarajah, Timothy Merritt

Socialization is an essential development skill for preschool children. In collaboration with the LEGO Group, we developed Robert Robot, a simplified robot, which enables socialization between children and facilitates shared experiences when meeting for the first time. An exploratory study to observe socialization between preschool children was conducted wit

Yuma Narita, Wen Yin

Recently, a high-energy neutrino event, designated KM3-230213A, was observed by the KM3NeT/ARCA detector in the Mediterranean Sea. This event is characterized by a reconstructed muon energy of approximately 120 PeV, corresponding to a median neutrino energy of roughly 220 PeV. To understand the origin, it is essential to investigate consistency with multi-me

Debabrota Basu, Debarshi Chanda

Resource-efficiently computing representations of probability distributions and the distances between them while only having access to the samples is a fundamental and useful problem across mathematical sciences. In this paper, we propose a generic framework to learn the probability and cumulative distribution functions (PDFs and CDFs) of a sub-Weibull, i.e.

Feng-Yuan Liu, James S. Dunlop, Ross J. McLure, Derek J. McLeod

We use the deep NIRCam and MIRI imaging from the JWST PRIMER survey to study the properties of (sub)mm sources detected by ALMA in the centre of the COSMOS field, with the aim of better constraining the history of dust-enshrouded star formation. The wealth of ALMA data in this field enabled us to isolate a robust sample of 128 (sub)mm sources within the 175

Mikhail V. Tamm

I overview data on several radical societal changes started circa 2015: accelerated decline in fertility rate, backsliding of democracy, rise of populist politics and arrest in generational renewal of political leadership. I conjecture that all these processes have a common underlying cause: the spread of cheap and easy access to information due to wide spre

Liwei Che, Tony Qingze Liu, Jing Jia, Weiyi Qin

Despite their remarkable potential, Large Vision-Language Models (LVLMs) still face challenges with object hallucination, a problem where their generated outputs mistakenly incorporate objects that do not actually exist. Although most works focus on addressing this issue within the language-model backbone, our work shifts the focus to the image input source,

Philipp Wu, Yide Shentu, Qiayuan Liao, Ding Jin

Learning from human demonstration is an effective approach for learning complex manipulation skills. However, existing approaches heavily focus on learning from passive human demonstration data for its simplicity in data collection. Interactive human teaching has appealing theoretical and practical properties, but they are not well supported by existing huma

José Gonçalves, Miguel Silva, Bernardo Cabral, Tiago Dias

Deep Learning (DL) has emerged as a powerful tool for vulnerability detection, often outperforming traditional solutions. However, developing effective DL models requires large amounts of real-world data, which can be difficult to obtain in sufficient quantities. To address this challenge, DiverseVul dataset has been curated as the largest dataset of vulnera

Antoine Legouhy, Ross Callaghan, Nolah Mazet, Vivien Julienne

This paper presents NimbleReg, a light-weight deep-learning (DL) framework for diffeomorphic image registration leveraging surface representation of multiple segmented anatomical regions. Deep learning has revolutionized image registration but most methods typically rely on cumbersome gridded representations, leading to hardware-intensive models. Reliable fi

Yehyun Suh, J. Ryan Martin, Daniel Moyer

This paper presents an approach for improving 2D/3D pelvis registration in optimization-based pose estimators using a learned initialization function. Current methods often fail to converge to the optimal solution when initialized naively. We find that even a coarse initializer greatly improves pose estimator accuracy, and improves overall computational effi

Badhan Kumar Das, Ajay Singh, Saahil Islam, Gengyan Zhao

The Transformer architecture has opened a new paradigm in the domain of deep learning with its ability to model long-range dependencies and capture global context and has outpaced the traditional Convolution Neural Networks (CNNs) in many aspects. However, applying Transformer models to 3D medical image datasets presents significant challenges due to their h

Brian Nelson, Behrouz Farhang-Boroujeny

The use of non-orthogonal signals has several benefits over orthogonal signals in multi-coded communications. We provide a novel, theoretical study of non-orthogonal signaling to expand the applicability of these schemes. Motivated by a class of multi-carrier spread spectrum systems, this paper presents a thorough symbol error rate analysis of the broad clas

Yehyun Suh, J. Ryan Martin, Daniel Moyer

This paper presents a novel method for estimating the orientation and the position of acetabular hip implants in total hip arthroplasty using full anterior-posterior hip fluoroscopy images. Our method accounts for distortions induced in the fluoroscope geometry, estimating acetabular component pose by creating a forward model of the perspective projection an

Jeel Chatrola, Abhiroop Ajith, Kevin Leahy, Constantinos Chamzas

We propose a novel, multi-layered planning approach for computing paths that satisfy both kinodynamic and spatiotemporal constraints. Our three-part framework first establishes potential sequences to meet spatial constraints, using them to calculate a geometric lead path. This path then guides an asymptotically optimal sampling-based kinodynamic planner, whi

Xinyi Liu, Ruijie Wang, Dachun Sun, Dilek Hakkani-Tur

Cross-Domain Recommendation (CDR) seeks to enhance item retrieval in low-resource domains by transferring knowledge from high-resource domains. While recent advancements in Large Language Models (LLMs) have demonstrated their potential in Recommender Systems (RS), their ability to effectively transfer domain knowledge for improved recommendations remains und

Xiao Zhang, Emmanouil Kioupakis

Rutile GeO2 is an emerging ultra-wide band gap semiconductor (UWBG) that has demonstrated excellent potential for applications in power electronic devices. Alloys of rutile SnO2, a well-established UWBG semiconducting oxide, with GeO2 are promising for tuning the material properties for applications. The thermal conductivity, in particular, is a key property

Linhan Fang, Fan Jiang, Ann Mary Toms, Xingpeng Li

Accurate wind speed prediction is crucial for designing and selecting sites for offshore wind farms. This paper investigates the effectiveness of various machine learning models in predicting offshore wind power for a site near the Gulf of Mexico by analyzing meteorological data. After collecting and preprocessing meteorological data, nine different input fe

Marco Pezzutto, Gabriele De Chiara, Stefano Gherardini

We determine the Kirkwood-Dirac quasiprobability (KDQ) distribution associated to the stochastic instances of internal energy variations for the quantum system and environment particles in coherent Markovian collision models. In the case the interactions between the quantum system and the particles do not conserve energy, the KDQ of the non-energy-preserving

Tomasz Romanczukiewicz, Yakov Shnir

We study long-term evolution of radiating quasi-Q-balls in 1+1 dimensional models without mass threshold. Two different models are considered, the model with a rational modification of the usual Q-ball sextic potential and the model of a Q-ball in a box with outgoing boundary conditions. We find that the outgoing boundary conditions modify the angular freque

Satoshi Funabashi, Atsumu Hiramoto, Naoya Chiba, Alexander Schmitz

To achieve a desired grasping posture (including object position and orientation), multi-finger motions need to be conducted according to the the current touch state. Specifically, when subtle changes happen during correcting the object state, not only proprioception but also tactile information from the entire hand can be beneficial. However, switching moti

Mariam Mughees, Yuzhuo Li, Yize Chen, Yunwei Ryan Li

Recent research shows large-scale AI-centric data centers could experience rapid fluctuations in power demand due to varying computation loads, such as sudden spikes from inference or interruption of training large language models (LLMs). As a consequence, such huge and fluctuating power demand pose significant challenges to both data center and power utilit

Fernando M. Belchior, Roberto V. Maluf, Carlos Alberto S. Almeida

In this work, we study a codimension-one thick brane within the framework of $f(Q,\mathcal{T})$ modified symmetric teleparallel gravity, where $Q$ is the nonmetricity scalar and $\mathcal{T}$ is the trace of the energy-momentum tensor. Using two specific choices of the warp factor, we construct the complete brane system, including the scalar field solution a

D. Turis-Gallo, M. Curé, R. S. Levenhagen, C. Arcos

The physical properties of stellar atmospheres in rapidly rotating massive stars, such as Be stars, are critical to understanding their evolution and their role as progenitors of supernovae. These stars, which often have near-critical rotation, exhibit equatorial stretching and gravity darkening, which significantly complicates the determination of parameter

Marcel Bernet, Pau Ramos, Teresa Antoja, Victor P. Debattista

The coupling between the dark matter (DM) halo and the stellar disc is a key factor in galactic evolution. While the interaction between structures like the Galactic bar and DM halos has been explored (e.g. slowing down of the bar due to dynamical friction), the effect of spiral arms on the DM halo distribution has received limited attention. We analyze a su

Francesco Capozzi, William Giarè, Eligio Lisi, Antonio Marrone

We perform an updated global analysis of the known and unknown parameters of the standard $3\nu$ framework as of 2025. The known oscillation parameters include three mixing angles $(\theta_{12},\,\theta_{23},\,\theta_{13})$ and two squared mass gaps, chosen as $\delta m^2=m^2_2-m^2_1>0$ and $\Delta m^2=m^2_3-{\textstyle\frac{1}{2}}(m^2_1+m^2_2)$, where $\alp

Alexander-Georg Penner, Ludmila Viotti, Rosario Fazio, Liliana Arrachea

We introduce a model of an active quantum particle and discuss its properties. The particle has a set of internal states that mediate exchanges of heat with external reservoirs. Heat is then converted into motion by means of a spin-orbit term that couples internal and translational degrees of freedom. The quantum features of the active particle manifest both

Shafiuddin Rehan Ahmed, Ankit Parag Shah, Quan Hung Tran, Vivek Khetan

Climate change has intensified the need for transparency and accountability in organizational practices, making Environmental, Social, and Governance (ESG) reporting increasingly crucial. Frameworks like the Global Reporting Initiative (GRI) and the new European Sustainability Reporting Standards (ESRS) aim to standardize ESG reporting, yet generating compre

Yuxuan Wang, Andrey V. Chubukov

In the BCS theory of superconductivity, an instability towards pairing develops at arbitrary weak dimensionless coupling $\lambda$ due to a divergence of logarithmic perturbative series for the pairing susceptibility (Cooper logarithms) at $T_c \sim \omega_0 e^{-1/\lambda}$, where $\omega_0$ is an energy cutoff. On the contrary, in many models of superconduc

Yukun Wang, Yingtong Shen, Zhichao Zhang, Linchun Wan

The graph isomorphism problem remains a fundamental challenge in computer science, driving the search for efficient decision algorithms. Due to its ambiguous computational complexity, heuristic approaches such as simulated annealing are frequently used, achieving high solution probabilities while avoiding exhaustive enumeration. However, traditional simulate

Cansu Korkmaz, Nancy Mehta, Radu Timofte

Recovering high-frequency details and textures from low-resolution images remains a fundamental challenge in super-resolution (SR), especially when real-world degradations are complex and unknown. While GAN-based methods enhance realism, they suffer from training instability and introduce unnatural artifacts. Diffusion models, though promising, demand excess

Xu Zhang, Nick Bultinck

A powerful way to guarantee the absence of a sign problem in determinantal quantum Monte Carlo simulations is imposing a particular type of anti-unitary symmetries. It is shown that these same symmetries give rise to constraints on correlation functions, which can be used to identify local operators whose correlations upper bound those of a large class of ph

Sergio Martínez-González

The feedback from massive stars drives the evolution of interstellar dust grains by altering their physical properties via a number of radiative and mechanical processes. Through these interactions, interstellar grains can achieve high rotational velocities due to unbalanced torques, potentially leading to their disruption. Mechanical torque disruption occur

Zachary Mann, Ningping Cao, Raymond Laflamme, Sisi Zhou

Quantum metrology aims to maximize measurement precision on quantum systems, with a wide range of applications in quantum sensing. Achieving the Heisenberg limit (HL) - the fundamental precision bound set by quantum mechanics - is often hindered by noise-induced decoherence, which typically reduces achievable precision to the standard quantum limit (SQL). Wh

Sayak Biswas, Saurav Suman, Mohit Randeria, Rajdeep Sensarma

STM experiments in the tunneling and Andreev regimes on graphene-based moire superconductors (SC) show two distinct energy scales whose origin is mysterious. We express the conductance of a normal-SC interface in terms of Green's functions, which allows us to sharpen the issues in two ways. First, we show that the two distinct energy scales cannot be underst

Michael Adlerstein, João Carlos Virgolino Soares, Angelo Bratta, Claudio Semini

Point cloud registration is a critical problem in computer vision and robotics, especially in the field of navigation. Current methods often fail when faced with high outlier rates or take a long time to converge to a suitable solution. In this work, we introduce a novel algorithm for point cloud registration called SANDRO (Splitting strategy for point cloud

Jun Qi Fang, Xiao Yan Xu

Over the past two decades, the overlap matrix approach has been developed to compute quantum entanglement in free-fermion systems, particularly to calculate entanglement entropy and entanglement negativity. This method involves the use of partial trace and partial transpose operations within the overlap matrix framework. However, in previous studies, only th

R. Hurtado-Gutiérrez, C. Pérez-Espigares, P. I. Hurtado

Time crystals are many-body systems whose ground state spontaneously breaks time-translation symmetry and thus exhibits long-range spatiotemporal order and robust periodic motion. Using hydrodynamics, we have recently shown how an $m$th-order external packing field coupled to density fluctuations in driven diffusive fluids can induce the spontaneous emergenc

A. de Oliveira Junior, Jonatan Bohr Brask, Rafael Chaves

The birth, life, and death of Maxwell's demon provoked a profound discussion about the interplay between thermodynamics, computation, and information. Even after its exorcism, the demon continues to inspire a multidisciplinary field. This tutorial offers a comprehensive overview of Maxwell's demon and its enduring influence, bridging classical concepts with

Cameron Smith, Basile Van Hoorick, Vitor Guizilini, Yue Wang

Motion serves as a powerful cue for scene perception and understanding by separating independently moving surfaces and organizing the physical world into distinct entities. We introduce SIRE, a self-supervised method for motion discovery of objects and dynamic scene reconstruction from casual scenes by learning intrinsic rigidity embeddings from videos. Our

Tawny Sit, David H. Weinberg, Emily J. Griffith

Using multi-element abundances from the SDSS APOGEE survey, we investigate the origin of abundance variations in Milky Way (MW) disk stars on the "high-$\alpha$ plateau," with $-0.5\leq\rm{[Mg/H]}\leq-0.1$ and $0.25\leq\rm{[Mg/Fe]}\leq0.35$. The elevated [$\alpha$/Fe] ratios of these stars imply low enrichment contributions from Type Ia supernovae (SNIa), bu

Shengfan Cao, Eunhyek Joa, Francesco Borrelli

Guaranteeing constraint satisfaction is challenging in imitation learning (IL), particularly in tasks that require operating near a system's handling limits. Traditional IL methods, such as Behavior Cloning (BC), often struggle to enforce constraints, leading to suboptimal performance in high-precision tasks. In this paper, we present a simple approach to in

Tiago P. Peixoto

Network reconstruction is the task of inferring the unseen interactions between elements of a system, based only on their behavior or dynamics. This inverse problem is in general ill-posed, and admits many solutions for the same observation. Nevertheless, the vast majority of statistical methods proposed for this task -- formulated as the inference of a grap

A. M. Wisłocka, J. Stücker, O. Hahn, R. E. Angulo

The $\Lambda$CDM model predicts structure formation across a vast mass range, from massive clusters ($\sim10^{15}\,\text{M}_\odot$) to Earth-mass micro-haloes ($\sim 10^{-6} \, \text{M}_\odot$), resolving which far exceeds the capabilities of current simulations. Excursion set models are the most efficient theoretical tool to disentangle this hierarchy in ma

Vladyslav Pushenko, Simone Scollo, Patrice Meunier, Emmanuel Villermaux

The mixing properties of vapor content, temperature and particle fields are of paramount importance in cloud turbulence as they pertain to essential processes, such as cloud water droplet evaporation and entrainment. Our study examines the mixing of a single cloudy air (which implies droplet-laden) filament with its clear air environment, a characteristic pr

S. Schuldt, R. Cañameras, Y. Shu, I. T. Andika

We have carried out a systematic search for galaxy-scale lenses exploiting multi-band imaging data from the third public data release of the Hyper Suprime-Cam (HSC) survey with the focus on false-positive removal, after applying deep learning classifiers to all 110 million sources with i-Kron radius above 0.8". To improve the performance, we tested the combi

Shivam Singh Kushwah, Aalok Misra

After obtaining the flavor $D6$-brane gauge fields/their fluctuations in the type IIA dual of T>T_c QCD-like theories at intermediate coupling (via the ${\cal M}$-theory uplift's ${\cal O}(R^4)$ corrections) in the absence/presence of a strong magnetic field, we compute the photoproduction spectral function and get a nice agreement with gauged supergravity b

Fo-Hong Wang, Xiao Yan Xu

Entanglement entropy is a fundamental measure of quantum entanglement for pure states, but for large-scale many-body systems, R\'{e}nyi entanglement entropy is much more computationally accessible. For mixed states, logarithmic negativity (LN) serves as a widely used entanglement measure, but its direct computation is often intractable, leaving R\'{e}nyi neg

Aly Haroon, Waleed Elsanhoury, Essam Elkholy, Abdel naby Saad

In this study, we utilize photometric and kinematic data from \textit{Gaia} DR3 and the {\sc ASteCA} package to analyze the sparsely studied open clusters, King 2 and King 5. For King 2, we identify 340 probable members with membership probabilities exceeding 50\%. Its mean proper motion components are determined as $(\mu_\alpha\cos\delta,~\mu_\delta) = (-1.

Amit Vikram

Eigenstate thermalization has played a prominent role as a determiner of the validity of quantum statistical mechanics since von Neumann's early works on quantum ergodicity. However, its connection to the dynamical process of quantum thermalization relies sensitively on nondegeneracy properties of the energy spectrum, as well as detailed features of indi

Xiaolong Du, Daniel Gilman, Tommaso Treu, Andrew Benson

Strong gravitational lenses enable direct inference of halo abundance and internal structure, which in turn enable constraints on the nature of dark matter and the primordial matter power spectrum. However, the density profiles of dark subhalos around the main deflector of a strong lens system also depend on tidal evolution inside the host, complicating the

Alonso Perez-Lona

We extend the decomposition conjecture to 2d quantum field theories with a gauged $\text{Rep}(H)$ symmetry category for $H$ a finite-dimensional semisimple Hopf algebra with $\text{Rep}(G)$ trivially-acting and $\text{Vec}(\Gamma)$ the remaining symmetry, for $G,\Gamma$ finite groups. We check our extension by explicitly computing partition functions, and by

Miguel Antonio Sulangi

Scanning tunneling spectroscopy experiments on a number of cuprate superconductors have revealed that these materials are highly inhomogeneous. However, even though this inhomogeneity is well-characterized experimentally, a theoretical understanding of the effect of an inhomogeneous superconducting $d$-wave order parameter on various observables is still not

A. G. Kim, P. E. Nugent, Xingzhuo Chen, L. Wang

This paper investigates the potential of intensity interferometry, based on the Hanbury Brown-Twiss effect, for measuring supernova sizes and distances. Through optimized telescope positioning, observing strategy, and advancements in single-photon detection technology, this method can provide precise angular size measurements of Type Ia supernovae as bright

Anke Biekötter, Benjamin D. Pecjak

We present analytic results for electroweak precision observables (EWPO) at next-to-leading order (NLO) in dimension-six SMEFT, with no assumptions on the flavour structure of SMEFT Wilson coefficients. The results are given in five different electroweak input schemes, thus offering a simple means, along with scale variations, of estimating theory uncertaint

Connor Basinger, Marshall C. Johnson, Ji Wang, Alison Duck

Ultra-hot Jupiters (UHJs) orbit close to their host stars and experience extreme conditions, making them important laboratories to explore atmospheric composition and dynamics. Transmission spectroscopy is a useful tool to reveal chemical species and their vertical and longitudinal distribution in the atmosphere. We use transmission spectra from the PEPSI sp

Julien Drapeau, Shreya Banerjee, Stefanos Kourtis

We introduce a variational algorithm based on the quantum alternating operator ansatz (QAOA) for the approximate solution of computationally hard counting problems. Our algorithm, dubbed VQCount, is based on the equivalence between random sampling and approximate counting and employs QAOA as a solution sampler. We first prove that VQCount improves upon previ

Calder Lenhart, Marshall C. Johnson, Ji Wang, Anusha Pai Asnodkar

KELT-20b is a well-studied ($T_{\text{eq}}=2262$ K) ultra hot Jupiter, but its multidimensional atmospheric structure remains unconstrained. We performed high-resolution cross-correlation transmission spectroscopy (HRCCTS) on a single transit time series of KELT-20b, observed with PEPSI on the LBT. Upon combining nineteen in-transit exposures, we detect Fe I

Kathrine Mørch Groth, Markus Ahlers

More than a decade ago, the IceCube Neutrino Observatory discovered a diffuse flux of 10 TeV-10 PeV neutrinos from our Universe. This flux of unknown origin most likely emanates from an extragalactic population of neutrino sources, which are individually too faint to appear as bright emitters. We review constraints on extragalactic neutrino source population

Jorge Sarrato-Alós, Christopher Brook, Arianna Di Cintio, Julen Expósito-Márquez

Determining the dynamical mass profiles of dispersion-supported galaxies is particularly challenging due to projection effects and the unknown shape of their velocity anisotropy profile. Our goal is to develop a machine learning algorithm capable of recovering dynamical mass profiles of dispersion-supported galaxies from line-of-sight stellar data. Tradition

Bo Xiao, Gonzalo Alvarez, Gábor B. Halász

Non-Abelian anyons in quantum spin liquids (QSLs) provide a promising route to fault-tolerant topological quantum computation. In the exactly solvable Kitaev honeycomb model, such anyons of the QSL state can be bound to nonmagnetic spin vacancies and endowed with non-Abelian statistics by an infinitesimal magnetic field. Here, we investigate how this approac

Nikki N. Geesink, Pavel E. Mancera Piña, Claudia del P. Lagos, Mariska Kriek

We present an analysis of the molecular specific angular momentum-mass ($j_{H_2}-M_{H_2}$) relation using a sample of 51 nearby disc galaxies from the PHANGS-ALMA survey with deep, high-resolution molecular gas rotation curves and surface density profiles. For the very first time, using a statistical sample, we report the discovery of a well-defined $j_{H_2}

Uri Keshet

Radio phoenixes are filamentary sources in the intracluster medium (ICM) of galaxy clusters, often extending over $>100$ kpc, arising from fossil radio lobes. Their soft, curved spectrum is widely attributed to aged relativistic electrons recently accelerated or compressed, but at high frequencies is shown to approach a power-law. Moreover, the full, curved

Dhashin Krishna, Rinchen Sherpa, Akash Kumar Saha, Tarak Nath Maity

Particle dark matter scattering on electrons in the Sun may gravitationally capture and self-annihilate inside it to neutrinos and anti-neutrinos, or other final states that in turn decay to them. Using up-to-date measurements by Super-Kamiokande of the fluxes of atmospheric electron-type and muon-type neutrinos, we set the most stringent limits on the elect

Fabian Pichler, Mohammad Hafezi, Michael Knap

Understanding interactions between excitons and correlated electronic states presents a fundamental challenge in quantum many-body physics. Here, we introduce a purely electronic model for the formation of exciton-polarons in moir\'e lattices. Unlike conventional approaches that treat excitons as tightly-bound bosonic particles, our model considers only elec

Fazeel Mahmood Khan, Benjamin L. Davis, Andrea Valerio Macciò, Kelly Holley-Bockelmann

Recent James Webb Space Telescope observations have revealed a peculiar class of galaxies at redshifts $z \gtrsim 6$, characterized by extremely high central stellar densities and overmassive central supermassive black holes (SMBHs), "little red dots" (LRDs). A critical question remains: If LRDs were common at high redshifts, how would they evolve into local

Jake Belton, Nadav Drukker, Ziwen Kong, Andreas Stergiou

The magnetic line defect in the $O(N)$ model gives rise to a non-trivial one-dimensional defect conformal field theory of theoretical and experimental value. This model is considered here in $d=4-\varepsilon$ and the full spectrum of defect operators with dimensions close to one, two and three at order $\varepsilon$ is presented. The spectrum of several clas

Rebecca Kyer, Subhroja Roy, Jay Strader, Ryan Urquhart

We report the discovery of two new Galactic accreting compact objects consistent with the respective positions of the unassociated Fermi-LAT gamma-ray sources 4FGL J0639.1--8009 and 4FGL J1824.2+1231. A combination of new and archival X-ray data from Chandra, XMM-Newton, Swift/XRT, and eROSITA reveals a variable X-ray source in each gamma-ray error ellipse.

Lei Gioia, Ryan Thorngren

We construct Hamiltonian models on a 3+1d cubic lattice for a single Weyl fermion and for a single Weyl doublet protected by exact (as opposed to emergent) chiral symmetries. In the former, we find a not-on-site, non-compact chiral symmetry which can be viewed as a Hamiltonian analog of the Ginsparg-Wilson symmetry in Euclidean lattice models of Weyl fermion

Giulio Salvatori

How to turn the flip of a coin into a random variable whose expected value equals a scattering amplitude? We answer this question by constructing a numerical algorithm to evaluate curve integrals - a novel formulation of scattering amplitudes - by a Monte Carlo strategy. To achieve a satisfactory accuracy we take advantage of tropical importance sampling. Th

S. Berta, G. Lagache, A. Beelen, R. Adam

(abridged) To understand early star formation, it is essential to determine the dust mass budget of high-redshift galaxies. Sub-millimeter rest-frame emission, dominated by cold dust, is an unbiased tracer of dust mass. The NIKA2 camera conducted a deep blank field survey at 1.2 and 2.0 mm in the GOODS-N field as part of the NIKA2 Cosmological Legacy Survey

Laura Shou, Alireza Parhizkar, Victor Galitski

We demonstrate the existence of transient two-dimensional surfaces where a random-walking particle escapes to infinity in contrast to localization in standard flat 2D space. We first prove that any rotationally symmetric 2D membrane embedded in flat 3D space cannot be transient. Then we formulate a criterion for the transience of a general asymmetric 2D memb

M. Veresvarska, S. Scaringi, C. Littlefield, D. de Martino

Magnetic accreting white dwarfs in cataclysmic variables have been known to show bursts driven by different physical mechanisms; however, the burst occurrence is much rarer than in their non-magnetic counterparts. DW Cnc is a well-studied intermediate polar that showed a burst with a 4-magnitude amplitude in 2007. Here we report on a recent burst in DW Cnc o

Anyuan Xu, Xiao Yan Chew, Dong-han Yeom

We investigate the dynamical collapse of Bronnikov-Ellis (BE) wormhole using the double-null formalism, where its throat is characterized by the coincidence of two curves $r_{,u} = 0$ and $r_{,v} = 0$. The emission of two ingoing pulses: normal scalar and phantom fields in the wormhole spacetime reveals two distinct instability scenarios: a normal scalar fie

Jacob Gildenblat, Jens Pahnke

We present Preserving Clusters and Correlations (PCC), a novel dimensionality reduction (DR) method a novel dimensionality reduction (DR) method that achieves state-of-the-art global structure (GS) preservation while maintaining competitive local structure (LS) preservation. It optimizes two objectives: a GS preservation objective that preserves an approxima

Bo Jiang, Shaoyu Chen, Qian Zhang, Wenyu Liu

OpenAI o1 and DeepSeek R1 achieve or even surpass human expert-level performance in complex domains like mathematics and science, with reinforcement learning (RL) and reasoning playing a crucial role. In autonomous driving, recent end-to-end models have greatly improved planning performance but still struggle with long-tailed problems due to limited common s

Ying Xu, Marius Pedersen, Kiran Raja

The rapid development of deep learning and generative AI technologies has profoundly transformed the digital contact landscape, creating realistic Deepfake that poses substantial challenges to public trust and digital media integrity. This paper introduces a novel Deepfake detention framework, Volume of Differences (VoD), designed to enhance detection accura

Sofiia Dubova, Kevin Yang, Horng-Tzer Yau, Jun Yin

We study a random band matrix $H=(H_{xy})_{x,y}$ of dimension $N\times N$ with mean-zero complex Gaussian entries, where $x,y$ belong to the discrete torus $(\mathbb{Z}/\sqrt{N}\mathbb{Z})^{2}$. The variance profile $\mathbb{E}|H_{xy}|^{2}=S_{xy}$ vanishes when the distance between $x,y$ is larger than some band-width parameter $W$ depending on $N$. We show

Xun Liang, Hanyu Wang, Huayi Lai, Simin Niu

Large Language Models have achieved remarkable success across various natural language processing tasks, yet their high computational cost during inference remains a major bottleneck. This paper introduces Sparse Expert Activation Pruning (SEAP), a training-free pruning method that selectively retains task-relevant parameters to reduce inference overhead. In

Lixue Gong, Xiaoxia Hou, Fanshi Li, Liang Li

Rapid advancement of diffusion models has catalyzed remarkable progress in the field of image generation. However, prevalent models such as Flux, SD3.5 and Midjourney, still grapple with issues like model bias, limited text rendering capabilities, and insufficient understanding of Chinese cultural nuances. To address these limitations, we present Seedream 2.

Tianhe Lin, Jian Xie, Siyu Yuan, Deqing Yang

Test-time compute is emerging as a new paradigm for enhancing language models' complex multi-step reasoning capabilities, as demonstrated by the success of OpenAI's o1 and o3, as well as DeepSeek's R1. Compared to explicit reasoning in test-time compute, implicit reasoning is more inference-efficient, requiring fewer generated tokens. However, why does the a

Sedrick Keh, Jean Mercat, Samir Yitzhak Gadre, Kushal Arora

Pre-trained LLMs that are further trained with image data perform well on vision-language tasks. While adding images during a second training phase effectively unlocks this capability, it is unclear how much of a gain or loss this two-step pipeline gives over VLMs which integrate images earlier into the training process. To investigate this, we train models

Yujie Wei, Shiwei Zhang, Hangjie Yuan, Biao Gong

Relational video customization refers to the creation of personalized videos that depict user-specified relations between two subjects, a crucial task for comprehending real-world visual content. While existing methods can personalize subject appearances and motions, they still struggle with complex relational video customization, where precise relational mo

Yuxin Jiang, Liming Jiang, Shuai Yang, Jia-Wei Liu

We present Style Matching Score (SMS), a novel optimization method for image stylization with diffusion models. Balancing effective style transfer with content preservation is a long-standing challenge. Unlike existing efforts, our method reframes image stylization as a style distribution matching problem. The target style distribution is estimated from off-

Rolf Pfister

The article analyses foundational principles relevant to the creation of artificial general intelligence (AGI). Intelligence is understood as the ability to create novel skills that allow to achieve goals under previously unknown conditions. To this end, intelligence utilises reasoning methods such as deduction, induction and abduction as well as other metho

Dünya Baradari, Nataliya Kosmyna, Oscar Petrov, Rebecah Kaplun

Generative AI is transforming education by enabling personalized, on-demand learning experiences. However, current AI systems lack awareness of the learner's cognitive state, limiting their adaptability. Meanwhile, electroencephalography (EEG)-based neuroadaptive systems have shown promise in enhancing engagement through real-time physiological feedback. Thi

Zeyinzi Jiang, Zhen Han, Chaojie Mao, Jingfeng Zhang

Diffusion Transformer has demonstrated powerful capability and scalability in generating high-quality images and videos. Further pursuing the unification of generation and editing tasks has yielded significant progress in the domain of image content creation. However, due to the intrinsic demands for consistency across both temporal and spatial dynamics, ach

Yuhong Zhang, Guanlin Wu, Ling-Hao Chen, Zhuokai Zhao

In this paper, we present a novel framework designed to reconstruct long-sequence 3D human motion in the world coordinates from in-the-wild videos with multiple shot transitions. Such long-sequence in-the-wild motions are highly valuable to applications such as motion generation and motion understanding, but are of great challenge to be recovered due to abru

Congyue Deng, Brandon Y. Feng, Cecilia Garraffo, Alan Garbarz

Machine learning frameworks for physical problems must capture and enforce physical constraints that preserve the structure of dynamical systems. Many existing approaches achieve this by integrating physical operators into neural networks. While these methods offer theoretical guarantees, they face two key limitations: (i) they primarily model local relation

Sinclair Schneider, Florian Steuber, Joao A. G. Schneider, Gabi Dreo Rodosek

The increasing popularity of large language models has not only led to widespread use but has also brought various risks, including the potential for systematically spreading fake news. Consequently, the development of classification systems such as DetectGPT has become vital. These detectors are vulnerable to evasion techniques, as demonstrated in an experi

Paul Mangold, Alain Durmus, Aymeric Dieuleveut, Eric Moulines

This paper proposes a novel analysis for the Scaffold algorithm, a popular method for dealing with data heterogeneity in federated learning. While its convergence in deterministic settings--where local control variates mitigate client drift--is well established, the impact of stochastic gradient updates on its performance is less understood. To address this

Youjun Zhao, Jiaying Lin, Rynson W. H. Lau

Open-vocabulary 3D object detection (OV-3DOD) aims at localizing and classifying novel objects beyond closed sets. The recent success of vision-language models (VLMs) has demonstrated their remarkable capabilities to understand open vocabularies. Existing works that leverage VLMs for 3D object detection (3DOD) generally resort to representations that lose th

A. Mironov, A. Morozov, A. Popolitov, Z. Zakirova

The triad refers to embedding of two systems of polynomials, symmetric ones and those of the Baker-Akhiezer type into a power series of the Noumi-Shiraishi type. It provides an alternative definition of Macdonald theory and its extensions. The basic triad is associated with the vector representation of the Ding-Iohara-Miki (DIM) algebra. We discuss lifting t

Bardia Safaei, Faizan Siddiqui, Jiacong Xu, Vishal M. Patel

Visual instruction tuning (VIT) for large vision-language models (LVLMs) requires training on expansive datasets of image-instruction pairs, which can be costly. Recent efforts in VIT data selection aim to select a small subset of high-quality image-instruction pairs, reducing VIT runtime while maintaining performance comparable to full-scale training. Howev

Anirban Chakraborty, Tirthankar Roy Choudhury

JWST has transformed our understanding of early galaxy formation, providing an unprecedented view of the first billion years of cosmic history. In this work, we build upon our previously developed semi-analytical framework that self-consistently models the evolving UVLF of galaxies and the global reionization history while incorporating the effects of radiat

Thomas Baxter, Shiho Kobayashi

Polarisation measurements of gamma-ray burst afterglows provide a powerful tool for probing the structure of relativistic jets. In this study, we revisit polarisation signals observed in gamma-ray burst afterglows, focusing on the effects of non-axisymmetric jet structures. To characterize these non-axisymmetric jets, we adopt a simple elliptical jet head mo

Junwei Luo, Yingying Zhang, Xue Yang, Kang Wu

Efficient vision-language understanding of large Remote Sensing Images (RSIs) is meaningful but challenging. Current Large Vision-Language Models (LVLMs) typically employ limited pre-defined grids to process images, leading to information loss when handling gigapixel RSIs. Conversely, using unlimited grids significantly increases computational costs. To pres

Dunant Cusipuma, David Ortega, Victor Flores-Benites, Arturo Deza

As multimodal foundational models start being deployed experimentally in Self-Driving cars, a reasonable question we ask ourselves is how similar to humans do these systems respond in certain driving situations -- especially those that are out-of-distribution? To study this, we create the Robusto-1 dataset that uses dashcam video data from Peru, a country wi

JaeWon Kim, Jiaying "Lizzy" Liu, Lindsay Popowski, Cassidy Pyle

Design has the potential to cultivate hope in the face of complex societal challenges. These challenges are often addressed through efforts aimed at harm reduction and prevention -- essential but sometimes limiting approaches that can unintentionally narrow our collective sense of what is possible. This one-day, in-person workshop builds on the first Positec

Paul C Bressloff

In this paper we explore the effects of instantaneous stochastic resetting on a planar slow-fast dynamical system of the form $\dot{x}=f(x)-y$ and $\dot{y}=\epsilon (x-y)$ with $0<\epsilon \ll 1$. We assume that only the fast variable $x(t)$ resets to its initial state $x_0$ at a random sequence of times generated from a Poisson process of rate $r$. Fixing t