March 2025 arXiv papers — page 28

Ishit Mehta, Manmohan Chandraker, Ravi Ramamoorthi

Problems in differentiable rendering often involve optimizing scene parameters that cause motion in image space. The gradients for such parameters tend to be sparse, leading to poor convergence. While existing methods address this sparsity through proxy gradients such as topological derivatives or lagrangian derivatives, they make simplifying assumptions abo

Carlo Bellavita, Eugenio Dellepiane, Georgios Stylogiannis

In this article we present new proofs for the boundedness and the compactness on $\ell^2$ of the Rhaly matrices, also known as terraced matrices. We completely characterize when such matrices belong to the Schatten class $\mathcal{S}^q(\ell^2)$, for $1<q<\infty$. Finally, we apply our results to study the Hadamard multipliers in weighted Dirichlet spaces, an

Tom Kempton, Stuart Burrell

Advances in hardware and language model architecture have spurred a revolution in natural language generation. However, autoregressive models compute probability distributions over next-token choices, and sampling from these distributions, known as decoding, has received significantly less attention than other design choices. Existing decoding strategies are

Ding Zhu, Zhiqun Zuo, Mohammad Mahdi Khalili

Large-scale machine learning (ML) models are increasingly being used in critical domains like education, lending, recruitment, healthcare, criminal justice, etc. However, the training, deployment, and utilization of these models demand substantial computational resources. To decrease computation and memory costs, machine learning models with sparse weight ma

Sahan Hewage Wewelwala, T. G. D. K. Sumanathilaka

This research presents a hybrid emotion recognition system integrating advanced Deep Learning, Natural Language Processing (NLP), and Large Language Models (LLMs) to analyze audio and textual data for enhancing customer interactions in contact centers. By combining acoustic features with textual sentiment analysis, the system achieves nuanced emotion detecti

Nandita Pan, Supratik Banerjee

Below critical temperature, turbulence prevents the spontaneous phase separation of binary mixtures, resulting in a phase arrested state of emulsion which is of significant interest for scientific and industrial applications. The current work is an extensive continuation of our initial theoretical investigation into the nature and universality of associated

Donghae Seo, Minyoung You, Hee-Cheol Kim, Gil Young Cho

Quantum Monte Carlo is a powerful tool for studying quantum many-body physics, yet its efficacy is often curtailed by the notorious sign problem. In this Letter, we introduce a novel criterion for the "intrinsic" sign problem in two-dimensional bosonic topological orders, which cannot be resolved by local basis transformations or adiabatic deformations of th

Zirui Wang, Tianying Wang

We consider the complex data modeling problem motivated by the zero-inflated and overdispersed data from microbiome studies. Analyzing how microbiome abundance is associated with human biological features, such as BMI, is of great importance for host health. Methods based on parametric distributional assumptions, such as zero-inflated Poisson and zero-inflat

Meng Wu

We establish several optimal estimates for exceptional parameters in the projection of fractal measures: (1) For a parametric family of self-similar measures satisfying a transversality condition, the set of parameters leading to a dimension drop is at most countable. (2) For any ergodic CP-distribution $Q$ on $\mathbb{R}^2$, the Hausdorff dimension of its o

R. O. Chametla, O. Chrenko, F. S. Masset, G. D'Angelo

Torques from asymmetric dust structures (so-called dust-void and filamentary structures) formed around low-mass planets embedded in a non turbulent dust-gas disk can exceed the torques produced by the gas disk component, then governing the planet's orbital dynamics. Here, we investigate how these structures (hence the dust torque) change when the effect of t

C. Marconcini, A. Feltre, I. Lamperti, M. Ceci

We present the analysis of the multi-phase gas properties in the Seyfert II galaxy NGC 424, using spatially resolved spectroscopic data from JWST/MIRI, part of the Mid-InfraRed Activity of Circumnuclear Line Emission (MIRACLE) program, as well as VLT/MUSE and ALMA. We trace the properties of the multi-phase medium, from cold and warm molecular gas to hot ion

Marco Ardu, Nicola Valori

The Electric Dipole Moment of the electron (eEDM) is typically investigated in experiments using paramagnetic molecules. However, the physical observable in these searches consists in a linear combination of CP-violating interactions, rather than the eEDM alone, which is commonly referred to as the equivalent EDM of the system. Assuming the presence of new C

Saleheh Ebadirad, Timothy W. Lyons, Gregory P. Fournier

This chapter explores the key carbon compounds that shaped the Archean biogeochemical cycle, delineating their substantial impact on Earth's primordial atmospheric and biospheric evolution. At the heart of the Archean carbon cycle were carbon dioxide and methane, which served as key regulators of Earth's early climate. Particular emphasis is placed on methan

Eduardo D. Stefanato, Nicolas M. Kawahala, Bianca A. Kawata, Paulo H. O. Rappl

Polar metals, materials that exhibit both electric polarization and high conductivity, can also host topological phases. Because free carriers strongly suppress distortive polar order and change the Fermi level, controlling charge dynamics is crucial for simultaneously tuning ferroelectric and topological phases in the same material. Here, we explore the exp

Marta Dell'Atti, Alessandra Rizzo, Pierandrea Vergallo

Led by the key example of the Korteweg-de Vries equation, we study pairs of Hamiltonian operators which are non-homogeneous and are given by the sum of a first-order operator and an ultralocal structure. We present a complete classification of the Casimir functions associated with the degenerate operators in two and three components. We define tensorial crit

Alexander Bastien, Omid Khormali

A graph $G$ is a link-irregular graph if every two distinct vertices of $G$ have non-isomorphic links. The link of a vertex $v$ in $G$ is the subgraph induced by the neighbors of $v$ in $G$. Ali, Chartrand and Zhang [Discussiones Mathematicae. Graph Theory, 45(1) (2025) p.95] conjectured that there exists no regular link-irregular graph. In this paper, we sh

A. V. Belitsky, V. A. Smirnov

We study the multi-collinear behavior of tree amplitudes in the six-dimensional N = (1,1) super Yang-Mills theory (sYM). A generalized dimensional reduction of the latter yields the four-dimensional N = 4 sYM on the Coulomb branch, which is of interest for considerations of massive or off-shell scattering. To this end, we revisit the calculation of tree scat

Johannes Seiffarth, Katharina Nöh

Tracking individual cells in live-cell imaging provides fundamental insights, inevitable for studying causes and consequences of phenotypic heterogeneity, responses to changing environmental conditions or stressors. Microbial cell tracking, characterized by stochastic cell movements and frequent cell divisions, remains a challenging task when imaging frame r

Dhrubajyoti Ghosh, Zhuolin Song, Luo Xiao, Sheng Luo

Accurately specifying covariance structures is critical for valid inference in longitudinal and functional data analysis, particularly when data are sparsely observed. In this study, we develop a global goodness-of-fit test to assess parametric covariance structures in multivariate sparse functional data. Our contribution is twofold. First, we extend the uni

Xiang Zheng, Anli Peng, Xi Hong, Cassidy R. Sugimoto

Interdisciplinary research has gained prominence as a necessity for addressing complex challenges, yet its impact on early academic careers remains unclear. This study examines how interdisciplinarity during doctoral training influences faculty placement at top universities across diverse fields. Analyzing the career trajectories of over 30,000 tenure-track

Sayed Muddashir Hossain, Simon Ostermann, Patrick Gebhard, Cord Benecke

Psychodynamic conflicts are persistent, often unconscious themes that shape a person's behaviour and experiences. Accurate diagnosis of psychodynamic conflicts is crucial for effective patient treatment and is commonly done via long, manually scored semi-structured interviews. Existing automated solutions for psychiatric diagnosis tend to focus on the recogn

Karima Kadaoui, Hanin Atwany, Hamdan Al-Ali, Abdelrahman Mohamed

We introduce JEEM, a benchmark designed to evaluate Vision-Language Models (VLMs) on visual understanding across four Arabic-speaking countries: Jordan, The Emirates, Egypt, and Morocco. JEEM includes the tasks of image captioning and visual question answering, and features culturally rich and regionally diverse content. This dataset aims to assess the abili

Abtin Daghighi

Brief Description: The book provides a unique highly self-contained text introducing the reader to the classical and modern theory of polyanalytic functions and their generalizations. This is a subbranch of complex analysis of several variables, in particular polyanalytic functions where introduced as a natural generalization of holomorphic functions. The bo

Carsten Henkel

Ultracold Bose gases are many-body systems with well-defined particle interactions that may serve as models for interacting quantum fields. The impact of virtual excitations is studied in the spatial transition zone created by a soft confinement potential that separates a degenerate ideal gas from a dense quasi-condensate. We compute within Bogoliubov theory

Yuying Duan, Gelei Xu, Yiyu Shi, Michael Lemmon

With the emerging application of Federated Learning (FL) in finance, hiring and healthcare, FL models are regulated to be fair, preventing disparities with respect to legally protected attributes such as race or gender. Two concepts of fairness are important in FL: global and local fairness. Global fairness addresses the disparity across the entire populatio

Oliver Heinimann, Assaf Shocher, Tal Zimbalist, Michal Irani

Traditional super-resolution (SR) methods assume an ``ideal'' downscaling SR-kernel (e.g., bicubic downscaling) between the high-resolution (HR) image and the low-resolution (LR) image. Such methods fail once the LR images are generated differently. Current blind-SR methods aim to remove this assumption, but are still fundamentally restricted to rather simpl

Anand Balakrishnan, Sheryl Paul, Simone Silvetti, Laura Nenzi

Modern cyber-physical systems (CPS) can consist of various networked components and agents interacting and communicating with each other. In the context of spatially distributed CPS, these connections can be dynamically dependent on the spatial configuration of the various components and agents. In these settings, robust monitoring of the distributed compone

Florent Ferro, Maurizio Fagotti

Quantum Fisher Information (QFI) is a ubiquitous quantity with applications ranging from quantum metrology and resource theories to condensed matter physics. In equilibrium local quantum many-body systems, the QFI of a subsystem with respect to an extensive observable is typically proportional to the subsystem's volume. Specifically, in large subsystems at e

Zhiwei Yang, Chen Gao, Jing Liu, Peng Wu

The rapid advancements in large language models (LLMs) have spurred growing interest in LLM-based video anomaly detection (VAD). However, existing approaches predominantly focus on video-level anomaly question answering or offline detection, ignoring the real-time nature essential for practical VAD applications. To bridge this gap and facilitate the practica

Martin Bourhis, Thomas Messmer, Michael Hölling, Oliver R. H. Buxton

This study investigates how variations in freestream turbulence (FST) and the thrust coefficient ($C_T$) influence wind turbine wakes. Wakes generated at $C_T \in \{0.5, 0.7,0.9\}$ are exposed to turbulent inflows with varying FST intensity ($1\% \lesssim TI_{\infty} \lesssim 11\%$) and integral length scale ($0.1 \lesssim \mathcal{L}_x/D \lesssim 2$, $D$ is

Hamed Babaei Giglou, Jennifer D'Souza, Oliver Karras, Sören Auer

Ontology Alignment (OA) is fundamental for achieving semantic interoperability across diverse knowledge systems. We present OntoAligner, a comprehensive, modular, and robust Python toolkit for ontology alignment, designed to address current limitations with existing tools faced by practitioners. Existing tools are limited in scalability, modularity, and ease

Ali Kogani, Behrooz Yousefzadeh

Nonreciprocity is most commonly associated with a large difference in the transmitted energy when the locations of the source and receiver are interchanged. This energy bias is accompanied by a difference in the transmitted phase. We highlight the role of this phase bias in breaking reciprocity in the steady-state vibration transmission characteristics of co

Shuai Li, Sree Sourav Das, Haobo Wang, Sujit Bati

Metallic thermoelectric materials are promising candidates for active cooling applications, where high thermal conductivity and a high thermoelectric power factor are essential to maximize effective thermal conductivity. While metals inherently possess high thermal and electrical conductivities, they typically exhibit low Seebeck coefficients. In this work,

Claudemir Alcantara, João Vitor da Silva, Ginaldo Sá

In this manuscript, we investigate geometric regularity estimates for problems governed by quasi-linear elliptic models in non-divergence form, which may exhibit either degenerate or singular behavior when the gradient vanishes, under strong absorption conditions of the form: \[ |\nabla u(x)|^{\gamma} \Delta_p^{\mathrm{N}} u(x) = f(x, u) \quad \text{in} \qua

Wei Zhong, Zhen-Tai Zhang, Bo Ma, Xianyu Tan

The non-grey picket-fence model predicts more accurately the temperatures in low-density regions compared to semi-grey models. This study investigates how the vertical mixing and convection fluxes modify the picket-fence model. The usual radiative-convective-equilibrium (RCE) is now extended to radiative-convective-mixing-equilibrium (RCME). The temperature

Areen Khalaila, Gianna Everette, Suho Kim, Ian Roy

Spatial reasoning and collaboration are essential for childhood development, yet blind and visually impaired (BVI) children often lack access to tools that foster these skills. Tactile maps and assistive technologies primarily focus on individual navigation, overlooking the need for playful, inclusive, and collaborative interactions. We address this with Str

Bangzheng Sun, Haojing Yan

We report an extremely tight, linear relation between ${\rm H\alpha}$ and [O~III] line fluxes in logarithm, discovered using a large sample of low and mid-resolution spectra (totaling 563) obtained by the James Webb Space Telescope (JWST) NIRSpec instrument in three widely separated extragalactic fields. While a certain correlation between ${\rm H\alpha}$ an

Leonardo Bettini, Bálint Kaszás, Bernhard Zybach, Jürg Dual

The dynamics in a primary Spectral Submanifold (SSM) constructed over the slowest modes of a dynamical system provide an ideal reduced-order model for nearby trajectories. Modeling the dynamics of trajectories further away from the primary SSM, however, is difficult if the linear part of the system exhibits strong non-normal behavior. Such non-normality impl

Charlie Cummings

We compute various averages over bulk geometries of quantum states prepared by the Chern-Simons path integral, for any level $k$ and compact simple gauge group $G$. We do so by carefully summing over all topologically distinct bulk geometries which have $n$ disjoint boundary tori and a decomposition into space$\times$time of fixed spatial topology. We find t

Taufiq Ahmed, Abhishek Kumar, Constantino Álvarez Casado, Anlan Zhang

Object detection models often struggle with class imbalance, where rare categories appear significantly less frequently than common ones. Existing sampling-based rebalancing strategies, such as Repeat Factor Sampling (RFS) and Instance-Aware Repeat Factor Sampling (IRFS), mitigate this issue by adjusting sample frequencies based on image and instance counts.

J. Ďurech, J. L. Ortiz, M. Ferrais, E. Jehin

Stellar occultations by asteroids observed from several stations are routinely used to reconstruct the asteroid's sky-plane projections. Together with the asteroid's shape model reconstructed from photometry, such occultations enable us to precisely determine its size and reveal details of its shape. When reducing occultation timings, the usual assumption is

Muhammad Abdullah Soomro, Adeel Nasrullah, Fatima Muhammad Anwar

Timekeeping is a fundamental component of modern computing; however, the security of system time remains an overlooked attack surface, leaving critical systems vulnerable to manipulation.

Kiana Burton, Meredith A. MacGregor, Rachel A. Osten, Ward S. Howard

Proxima Centauri (Cen) has been the subject of many flaring studies due to its proximity and potential to host habitable planets. The discovery of millimeter flares from this M dwarf with ALMA has opened a new window into the flaring process and the space-weather environments of exoplanets like Proxima b. Using a total of ~50 hours of ALMA observations of Pr

Patrice Bechard, Chao Wang, Amirhossein Abaskohi, Juan Rodriguez

Workflows are a fundamental component of automation in enterprise platforms, enabling the orchestration of tasks, data processing, and system integrations. Despite being widely used, building workflows can be complex, often requiring manual configuration through low-code platforms or visual programming tools. To simplify this process, we explore the use of g

Zeyad Alghamdi, Tharindu Kumarage, Garima Agrawal, Mansooreh Karami

Effective mental health support is crucial for alleviating psychological distress. While large language model (LLM)-based assistants have shown promise in mental health interventions, existing research often defines "effective" support primarily in terms of empathetic acknowledgments, overlooking other essential dimensions such as informational guidance, com

Wasim Akram, Manika Bag, Manil T. Mohan

In this article, we investigate a generalized Burgers-Huxley equation with a smooth kernel defined in a bounded domain $\Omega\subset\mathbb{R}^d$, $d\in\{1,2,3\}$, focusing on feedback stabilizability around a non-constant steady state. Initially, employing the Banach fixed point theorem, we establish the local existence and uniqueness of a strong solution,

Pilseo Park, Ze Zhang, Michel Sarkis, Ning Bi

High-fidelity reconstruction of head avatars from monocular videos is highly desirable for virtual human applications, but it remains a challenge in the fields of computer graphics and computer vision. In this paper, we propose a two-phase head avatar reconstruction network that incorporates a refined 3D mesh representation. Our approach, in contrast to exis

Bo-nan Zhang, Di Wang

Motivated by the first observation of $CP$ violation in baryon decays, we study the relations for $CP$ asymmetries of bottom baryon decays. It is found that all the $CP$ violation ratios of two $U$-spin conjugate decay channels satisfy a simple equation in the $U$-spin limit. As examples, we present the $U$-spin conjugate channels in the four-body decays of

Phillip C. Burke, Shane Dooley

A quantum many-body scar is an eigenstate of a chaotic many-body Hamiltonian that exhibits two seemingly incongruous properties: its energy eigenvalue corresponds to a high temperature, yet its entanglement structure resembles that of low-temperature eigenstates, such as ground states. Traditionally, a temperature is assigned to an energy \emph{eigenvalue} t

José Giral-Barajas, Paul C Bressloff

We study the accumulation of resources within a target due to the interplay between continual delivery, driven by 1D stochastic search processes, and sequential consumption. The assumption of sequential consumption is key because it changes the commonly used $G/M/\infty$ queue to a $G/M/c$ queue. Combining the theory of $G/M/c$ queues with the theory of firs

James Belk, Francesco Fournier-Facio, James Hyde, Matthew C. B. Zaremsky

We prove that the groups $\mathrm{Aut}(F_n)$ satisfy the Boone-Higman conjecture for all $n$, meaning each $\mathrm{Aut}(F_n)$ embeds in a finitely presented simple group. In fact, we prove that each $\mathrm{Aut}(F_n)$ satisfies the "permutational" Boone-Higman conjecture, which means the simple group in question can be taken to be a twisted Brin-Thompson g

Gil R. Cavalcanti, Jaime Pedregal, Roberto Rubio

We consider a family of metric generalized connections on transitive Courant algebroids, which includes the canonical Levi-Civita connection, and study the flatness condition. We find that the building blocks for such flat transitive Courant algebroids are compact simple Lie groups. Further, we give a description of left-invariant flat Levi-Civita generalize

Supritha Bhowmick, Mang Hei Gordon Lee, Diptimoy Ghosh, Farman Ullah

In this work we perform a systematic study of the singularity structure of inflationary correlations at 1-loop. We explicitly compute a few diagrams and find a pattern emerging in the singularities produced. Motivated by this, we derive diagrammatic rules to extract the singularities of any two-site 1-loop diagram. Using these rules, the poles and branch cut

Antti Rantala, Thorsten Naab

Motivated by JWST observations of dense, clumpy and clustered high redshift star formation, we simulate the hierarchical assembly of nine $M_{\mathrm{cl}}=10^6 M_\odot$ star clusters using the BIFROST N-body code. Our low metallicity models ($0.01Z_\odot$) with post-Newtonian equations of motion for black holes include evolving populations of single, binary

Audrey Huang, Adam Block, Qinghua Liu, Nan Jiang

Inference-time computation offers a powerful axis for scaling the performance of language models. However, naively increasing computation in techniques like Best-of-N sampling can lead to performance degradation due to reward hacking. Toward a theoretical understanding of how to best leverage additional computation, we focus on inference-time alignment, whic

Ayan Chakraborty, Simon Clery, Md Riajul Haque, Debaprasad Maity

We investigate the spectral behavior of scalar fluctuations generated by gravity during inflation and the subsequent reheating phase. We consider a non-perturbative Bogoliubov treatment within the context of pure gravitational reheating. We compute both long and short-wavelength spectra, first for a massless scalar field, revealing that the spectral index in

Edward Zhang, Andreas L. Faisst, Brendan Crill, Hanae Inami

Together with gas, stars, and supermassive black holes, dust is crucial in stellar and galaxy evolution. Hence, understanding galaxies' dust properties across cosmic time is critical to studying their evolution. In addition to photometric constraints on the absorption of blue light and its reemission at infrared wavelengths, dust grain properties can be expl

Aysha Aamer, Matt Nicholl, Sebastian Gomez, Edo Berger

Hydrogen-poor superluminous supernovae (SLSNe) are among the most energetic explosions in the universe, reaching luminosities up to 100 times greater than those of normal supernovae. Detailed spectral analysis hold the potential to reveal their progenitors and underlying energy sources. This paper presents the largest compilation of SLSN photospheric spectra

Rudolf Smolka, Jan Vysoky

Graded vector bundles over a given $\mathbb{Z}$-graded manifold can be defined in three different ways: certain sheaves of graded modules over the structure sheaf of the base graded manifold, finitely generated projective graded modules over the algebra of global functions on the base graded manifold, or locally trivial graded manifolds with a suitable linea

Gregg Musiker

We provide a superalgebraic analogue of Markov numbers, which are defined as the Grassmann integer solutions to the equation $x^2 + y^2 + z^2 + (xy + yz + xz)\epsilon = 3(1 + \epsilon)xyz$, as well as applications to the Decorated Super Teichm\"uller spaces associated to the once-punctured torus and certain annuli. We conclude with further directions for stu

Jonas El Gammal, Riccardo Buscicchio, Germano Nardini, Jesús Torrado

Source inference for deterministic gravitational waves is a computationally demanding task in LISA. In a novel approach, we investigate the capability of Gaussian Processes to learn the posterior surface in order to reconstruct individual signal posteriors. We use GPry, which automates this reconstruction through active learning, using a very small number of

Sangjun Cha, Boseong Young Cho, Hyungjin Joo, Wonki Lee

The Bullet Cluster (1E 0657-56) is a key astrophysical laboratory for studying dark matter, galaxy cluster mergers, and shock propagation in extreme environments. Using new JWST imaging, we present the highest-resolution mass reconstruction to date, combining 146 strong lensing constraints from 37 systems with high-density (398 sources arcmin$^{-2}$) weak le

R. J. Smethurst, B. D. Simmons, T. Géron, H. Dickinson

We have not yet observed the epoch at which disc galaxies emerge in the Universe. While high-$z$ measurements of large-scale features such as bars and spiral arms trace the evolution of disc galaxies, such methods cannot directly quantify featureless discs in the early Universe. Here we identify a substantial population of apparently featureless disc galaxie

Arjun Srinivasan Kudinoor, Daniel Pablos, Krishna Rajagopal

The ATLAS Collaboration has developed a method to analyze large-radius jets composed of skinny $R=0.2$ subjets in heavy-ion collisions. We first demonstrate that the measurements pioneered by ATLAS constrain the value of $L_{\rm res}$, the resolution length of QGP -- and rule out any picture in which an entire parton shower loses energy coherently as a singl

Gabriel Cuomo, Eren Firat, Filippo Nardi, Lorenzo Ricci

We study energy correlators and other event shapes in states created by operators with large global $U(1)$ charge $Q$ in Conformal Field Theories. Focusing on theories whose large charge sector is described by the superfluid Effective Field Theory (EFT), we develop a systematic framework to compute event shapes within the EFT. As formerly observed, event sha

Terry Generet, Kyle Lee, Ian Moult, Rene Poncelet

The study of hadronic jets and their substructure at hadronic colliders is crucial for improving our understanding of QCD, and searching for new physics. As such, there has been a significant effort to improve their theoretical description. In the small radius limit, inclusive jet production exhibits a universal factorization, enabling the resummation of log

Christian Capanelli, Wayne Hu, Evan McDonough

In the Fuzzy Dark Matter (FDM) scenario, the dark matter is composed of an ultra-light scalar field with coherence length and wave interference on astrophysical scales. Scalar fields generically have quartic self-interactions that modify their dispersion relation and the associated evolution of density perturbations. We perform the first dedicated analysis o

Jay Desai, Diptimoy Ghosh

In this paper, we attempt to derive ``positivity" bounds on Photon and Gluon Effective Field Theories (EFTs) at one loop level. While for the Photon case, the one loop amplitude is IR finite and well defined in the forward limit, earlier studies failed to obtain a dispersive bound on dimension-12 operators due to the dependence of the ``arc integral" on the

A. Lola Danhaive, Sandro Tacchella, Hannah Übler, Anna de Graaff

Recent studies of gas kinematics at high redshift have reported disky systems which appear to challenge models of galaxy formation, but it is unclear whether they are representative of the underlying galaxy population. We present the first statistical sample of spatially resolved ionised gas kinematics at high redshift, comprised of $272$ H$\alpha$ emitters

Ling Lin, Daniel Robbins, Subham Roy

We investigate the interplay between (-1)-form symmetries and their quantum-dual (d-1)-form counterparts within the framework of Symmetry Topological Field Theories (SymTFTs). In this framework the phenomenon of decomposition -- a d-dimensional quantum field theory with (d-1)-form symmetry being the disjoint union of other theories (or ''universes'') -- aris

Kang Yang, Zhi Li, Peng Xue, Emil J. Bergholtz

Topological phase transitions are typically characterized by abrupt changes in a quantized invariant. Here we report a contrasting paradigm in non-Hermitian parity-time symmetric systems, where the topological invariant remains conserved, but its nature transitions between the Chern number, characteristic of chiral transport in complex bands, and the Euler n

Abdelrahman Shaker, Muhammad Maaz, Chenhui Gou, Hamid Rezatofighi

Video understanding models often struggle with high computational requirements, extensive parameter counts, and slow inference speed, making them inefficient for practical use. To tackle these challenges, we propose Mobile-VideoGPT, an efficient multimodal framework designed to operate with fewer than a billion parameters. Unlike traditional video large mult

Chi-Pin Huang, Yen-Siang Wu, Hung-Kai Chung, Kai-Po Chang

Customized text-to-video generation aims to produce high-quality videos that incorporate user-specified subject identities or motion patterns. However, existing methods mainly focus on personalizing a single concept, either subject identity or motion pattern, limiting their effectiveness for multiple subjects with the desired motion patterns. To tackle this

Reza Qorbani, Gianluca Villani, Theodoros Panagiotakopoulos, Marc Botet Colomer

Open-vocabulary semantic segmentation models associate vision and text to label pixels from an undefined set of classes using textual queries, providing versatile performance on novel datasets. However, large shifts between training and test domains degrade their performance, requiring fine-tuning for effective real-world applications. We introduce Semantic

Weihao Yu, Yuanhao Cai, Ruyi Zha, Zhiwen Fan

Four-dimensional computed tomography (4D CT) reconstruction is crucial for capturing dynamic anatomical changes but faces inherent limitations from conventional phase-binning workflows. Current methods discretize temporal resolution into fixed phases with respiratory gating devices, introducing motion misalignment and restricting clinical practicality. In th

Ziren Gong, Fabio Tosi, Youmin Zhang, Stefano Mattoccia

NeRF-based SLAM has recently achieved promising results in tracking and reconstruction. However, existing methods face challenges in providing sufficient scene representation, capturing structural information, and maintaining global consistency in scenes emerging significant movement or being forgotten. To this end, we present HS-SLAM to tackle these problem

Jiahao Xie, Alessio Tonioni, Nathalie Rauschmayr, Federico Tombari

Visual in-context learning (VICL), as a new paradigm in computer vision, allows the model to rapidly adapt to various tasks with only a handful of prompts and examples. While effective, the existing VICL paradigm exhibits poor generalizability under distribution shifts. In this work, we propose test-time Visual In-Context Tuning (VICT), a method that can ada

Kaituo Feng, Kaixiong Gong, Bohao Li, Zonghao Guo

Inspired by DeepSeek-R1's success in eliciting reasoning abilities through rule-based reinforcement learning (RL), we introduce Video-R1 as the first attempt to systematically explore the R1 paradigm for incentivizing video reasoning within multimodal large language models (MLLMs). However, directly applying RL training with the GRPO algorithm to video reaso

Ziyu Guo, Young Yoon Lee, Joseph Liu, Yizhak Ben-Shabat

We present StyleMotif, a novel Stylized Motion Latent Diffusion model, generating motion conditioned on both content and style from multiple modalities. Unlike existing approaches that either focus on generating diverse motion content or transferring style from sequences, StyleMotif seamlessly synthesizes motion across a wide range of content while incorpora

Jianning Pei, Han Hu, Shuyang Gu

Diffusion models achieve remarkable generation quality but suffer from computational intensive sampling due to suboptimal step discretization. While existing works focus on optimizing denoising directions, we address the principled design of stepsize schedules. This paper proposes Optimal Stepsize Distillation, a dynamic programming framework that extracts t

Christophe Piveteau

It is one of the most fundamental objectives in quantum information science to understand the boundary between the computational power of classical and quantum computers. One possible avenue to explore this boundary is to identify classes of quantum circuits that can be efficiently simulated on a classical computer. Instead of simulating a general quantum ci

Zilin Xiao, Pavel Suma, Ayush Sachdeva, Hao-Jen Wang

We introduce LOCORE, Long-Context Re-ranker, a model that takes as input local descriptors corresponding to an image query and a list of gallery images and outputs similarity scores between the query and each gallery image. This model is used for image retrieval, where typically a first ranking is performed with an efficient similarity measure, and then a sh

Hongkai Lin, Dingkang Liang, Zhenghao Qi, Xiang Bai

Underwater dense prediction, especially depth estimation and semantic segmentation, is crucial for gaining a comprehensive understanding of underwater scenes. Nevertheless, high-quality and large-scale underwater datasets with dense annotations remain scarce because of the complex environment and the exorbitant data collection costs. This paper proposes a un

Anand Bhattad, Konpat Preechakul, Alexei A. Efros

This paper proposes a novel scene understanding task called Visual Jenga. Drawing inspiration from the game Jenga, the proposed task involves progressively removing objects from a single image until only the background remains. Just as Jenga players must understand structural dependencies to maintain tower stability, our task reveals the intrinsic relationsh

Fang Xie, Chenyuan Li, Jennifer Cano, Qimiao Si

The discovery of superconductivity and correlated electronic phases in twisted bilayer WSe$_2$ (Xia et al., Nature 2024; Guo et al., Nature 2025) has generated considerable excitement. Accompanying the superconductivity and a correlated insulator phase is the Kondo-lattice-like phenomenology in transport properties. Here we consider how such phenomenology ca

Daniela Bertacchi, Fabio Zucca

Germ order is a non-standard stochastic order defined through the comparison of the generating functions of the processes. This order was first introduced for branching random walks with a constant breeding law and independent dispersal of offspring, which are characterized by a one-dimensional generating function. In this work, we investigate the properties

Hairong Yin, Huangying Zhan, Yi Xu, Raymond A. Yeh

Open-vocabulary 3D scene understanding is crucial for robotics applications, such as natural language-driven manipulation, human-robot interaction, and autonomous navigation. Existing methods for querying 3D Gaussian Splatting often struggle with inconsistent 2D mask supervision and lack a robust 3D point-level retrieval mechanism. In this work, (i) we prese

Haolin Liu, Xiaohang Zhan, Zizheng Yan, Zhongjin Luo

Establishing character shape correspondence is a critical and fundamental task in computer vision and graphics, with diverse applications including re-topology, attribute transfer, and shape interpolation. Current dominant functional map methods, while effective in controlled scenarios, struggle in real situations with more complex challenges such as non-iso

Minghui Lin, Xiang Wang, Yishan Wang, Shu Wang

Recent advancements in video generation have witnessed significant progress, especially with the rapid advancement of diffusion models. Despite this, their deficiencies in physical cognition have gradually received widespread attention - generated content often violates the fundamental laws of physics, falling into the dilemma of ''visual realism but physica

Ömer M. Aksoy, Xiao-Gang Wen

We explore 1+1 dimensional (1+1D) gapped phases in systems with non-invertible symmetries, focusing on symmetry-protected topological (SPT) phases (defined as gapped phases with non-degenerate ground states), as well as SPT orders (defined as the differences between gapped/gapless phases with identical bulk excitations spectrum). For group-like symmetries, d

Fei Tang, Haolei Xu, Hang Zhang, Siqi Chen

Graphical User Interface (GUI) Agents have emerged as a transformative paradigm in human-computer interaction, evolving from rule-based automation scripts to sophisticated AI-driven systems capable of understanding and executing complex interface operations. This survey provides a comprehensive examination of the rapidly advancing field of LLM-based GUI Agen

Koki Fusejima, Takuya Ishihara

This paper provides a new approach for identifying and estimating the Average Treatment Effect on the Treated under a linear factor model that allows for multiple time-varying unobservables. Unlike the majority of the literature on treatment effects in linear factor models, our approach does not require the number of pre-treatment periods to go to infinity t

David Yifan Yao, Albert J. Zhai, Shenlong Wang

This paper presents a unified approach to understanding dynamic scenes from casual videos. Large pretrained vision foundation models, such as vision-language, video depth prediction, motion tracking, and segmentation models, offer promising capabilities. However, training a single model for comprehensive 4D understanding remains challenging. We introduce Uni

Rana Salama, Jason Cai, Michelle Yuan, Anna Currey

Large language model (LLM) agents have evolved to intelligently process information, make decisions, and interact with users or tools. A key capability is the integration of long-term memory capabilities, enabling these agents to draw upon historical interactions and knowledge. However, the growing memory size and need for semantic structuring pose significa

Rita Tojeiro, Katarina Kraljic

The formation and evolution of galaxies cannot be separated from large scale structure growth. Dark matter halos (and, therefore, galaxies) form and grow within the cosmic web - the classification of large-scale structure as distinct environments, namely voids, walls, filaments and nodes. Thanks to the rapid development of extragalactic spectroscopic redshif

Qi Qin, Le Zhuo, Yi Xin, Ruoyi Du

We introduce Lumina-Image 2.0, an advanced text-to-image generation framework that achieves significant progress compared to previous work, Lumina-Next. Lumina-Image 2.0 is built upon two key principles: (1) Unification - it adopts a unified architecture (Unified Next-DiT) that treats text and image tokens as a joint sequence, enabling natural cross-modal in

Adrian Bulat, Yassine Ouali, Georgios Tzimiropoulos

In this work, we aim to compress the vision tokens of a Large Vision Language Model (LVLM) into a representation that is simultaneously suitable for (a) generative and (b) discriminative tasks, (c) is nearly lossless, and (d) is storage-efficient. We propose a novel compression approach, called Fwd2Bot, that uses the LVLM itself to compress the visual inform

Minyoung Kim

The bridge problem is to find an SDE (or sometimes an ODE) that bridges two given distributions. The application areas of the bridge problem are enormous, among which the recent generative modeling (e.g., conditional or unconditional image generation) is the most popular. Also the famous Schr\"{o}dinger bridge problem, a widely known problem for a century, i

Dian Zheng, Ziqi Huang, Hongbo Liu, Kai Zou

Video generation has advanced significantly, evolving from producing unrealistic outputs to generating videos that appear visually convincing and temporally coherent. To evaluate these video generative models, benchmarks such as VBench have been developed to assess their faithfulness, measuring factors like per-frame aesthetics, temporal consistency, and bas

Alessandro De Stefani, Eloísa Grifo, Jack Jeffries

We characterize symbolic powers of prime ideals in polynomial rings over any field in terms of $\mathbb{Z}$-linear differential operators, and of prime ideals in polynomial rings over complete discrete valuation rings with a $p$-derivation $\delta$ in terms of $\mathbb{Z}$-linear differential operators and of $\delta$. This extends previous work of the same