March 2023 arXiv papers — page 70

Mustafa Shukor, Corentin Dancette, Matthieu Cord

Large Language Models (LLMs) have so far impressed the world, with unprecedented capabilities that emerge in models at large scales. On the vision side, transformer models (i.e., ViT) are following the same trend, achieving the best performance on challenging benchmarks. With the abundance of such unimodal models, a natural question arises; do we need also t

Sayar Karmakar, Moumanti Podder, Souvik Roy, Soumyarup Sadhukhan

We study the bond percolation game and the site percolation game on the rooted Galton-Watson tree $T_{\chi}$ with offspring distribution $\chi$. We obtain the probabilities of win, loss and draw for each player in terms of the fixed points of functions that involve the probability generating function $G$ of $\chi$, and the parameters $p$ and $q$. Here, $p$ i

Leonardo Rodrigues Cadorim, Lucas Veneziani de Toledo, Wilson Aires Ortiz, Jorge Berger

By using the full 3D generalized time dependent Ginzbug-Landau equation we study a long superconducting film of finite width and thickness under an applied transport current. We show that, for sufficiently large thickness, the vortices and the antivortices become curved before they annihilate each other. As they approach the center of the sample, their ends

Simon Morelli, Christopher Eltschka, Marcus Huber, Jens Siewert

We present a novel inequality on the purity of a bipartite state depending solely on the difference of the local Bloch vector lengths. For two qubits this inequality is tight for all marginal states and so extends the previously known solution for the 2-qubit marginal problem and opens a new research avenue. We further use this inequality to construct a 3-di

Apoorva Lal, Mac Lockhart, Yiqing Xu, Ziwen Zu

Instrumental variable (IV) strategies are widely used in political science to establish causal relationships. However, the identifying assumptions required by an IV design are demanding, and it remains challenging for researchers to assess their validity. In this paper, we replicate 67 papers published in three top journals in political science during 2010-2

Mark E. AlSukaiti, Nafaa Chbili

We establish a relationship between the Jones polynomial of generalized weaving knots of type $W(3,n,m)$ and the Chebyshev polynomial of the first kind. Consequently, we prove that the coefficients of the Jones polynomial of weaving knots are basically the Whitney numbers of Lucas lattices. Furthermore, we give an explicit formula for the Alexander polynomia

Naveen A. Reddy, Michael W. Topping, Ryan L. Sanders, Alice E. Shapley

We conduct a statistical analysis of the factors responsible for the variation in the ionization parameter (U) of high-redshift star-forming galaxies based on medium resolution JWST/NIRSpec observations obtained by the Cosmic Evolution Early Release Science (CEERS) survey. The sample consists of 48 galaxies with spectroscopic redshifts z=2.7-6.3 which are la

Dave Zhenyu Chen, Yawar Siddiqui, Hsin-Ying Lee, Sergey Tulyakov

We present Text2Tex, a novel method for generating high-quality textures for 3D meshes from the given text prompts. Our method incorporates inpainting into a pre-trained depth-aware image diffusion model to progressively synthesize high resolution partial textures from multiple viewpoints. To avoid accumulating inconsistent and stretched artifacts across vie

Melissa Diamond, Damiano F. G. Fiorillo, Gustavo Marques-Tavares, Edoardo Vitagliano

New feebly interacting particles would emerge from a supernova core with 100-MeV-range energies and produce $\gamma$-rays by subsequent decays. These would contribute to the diffuse cosmic $\gamma$-ray background or would have shown up in the Solar Maximum Mission (SMM) satellite from SN~1987A. However, we show for the example of axion-like particles (ALPs)

Virginia Bresci, Martin Lemoine, Laurent Gremillet

The efficiency of particle acceleration at shock waves in relativistic, magnetized astrophysical outflows is a debated topic with far-reaching implications. Here, for the first time, we study the impact of turbulence in the pre-shock plasma. Our simulations demonstrate that, for a mildly relativistic, magnetized pair shock (Lorentz factor $\gamma_{\rm sh} \s

Richard J. Parker, Tim Lichtenberg, Miti Patel, Cheyenne K. M. Polius

The abundance of the short-lived radioisotopes 26-Al and 60-Fe in the early Solar system is usually explained by the Sun either forming from pre-enriched material, or the Sun's protosolar disc being polluted by a nearby supernova explosion from a massive star. Both hypotheses suffer from significant drawbacks: the former does not account for the dynamical ev

Ben Gillen, Jonathan Michala

Rothe diagrams are diagrams which track inversions of a permutation. We define six main properties that Rothe diagrams fulfill: the southwest, dot, popping, numbering, step-out avoiding, and empty cell gap rules. We prove that -- given an arbitrary bubble diagram -- four different subsets of these properties provide sufficient criteria for the diagram to be

R. Loubere, E. Macca, C. Pares, G. Russo

In this paper we blend high-order Compact Approximate Taylor (CAT) numerical methods with the a posteriori Multi-dimensional Optimal Order Detection (MOOD) paradigm to solve hyperbolic systems of conservation laws. The resulting methods are highly accurate for smooth solutions, essentially non-oscillatory for discontinuous ones, and almost fail-safe positivi

Max Peter Ronecker, Michael Stolz, Daniel Watzenig

Through constant improvements in recent years radar sensors have become a viable alternative to lidar as the main distancing sensor of an autonomous vehicle. Although robust and with the possibility to directly measure the radial velocity, it brings it's own set of challenges, for which existing algorithms need to be adapted. One core algorithm of a percepti

Fabio A. Faria, Luiz H. Buris, Luis A. M. Pereira, Fábio A. M. Cappabianco

Aerial scene classification, which aims to semantically label remote sensing images in a set of predefined classes (e.g., agricultural, beach, and harbor), is a very challenging task in remote sensing due to high intra-class variability and the different scales and orientations of the objects present in the dataset images. In remote sensing area, the use of

Kwun Chuen Gary Chan, Hok Kan Ling, Chuan-Fa Tang, Sheung Chi Phillip Yam

In this article, we propose a new class of consistent tests for $p$-variate normality. These tests are based on the characterization of the standard multivariate normal distribution, that the Hessian of the corresponding cumulant generating function is identical to the $p\times p$ identity matrix and the idea of decomposing the information from the joint dis

Shashank Saxena, Jan-Hendrik Bastek, Miguel Spinola, Prateek Gupta

Overcoming the time scale limitations of atomistics can be achieved by switching from the state-space representation of Molecular Dynamics (MD) to a statistical-mechanics-based representation in phase space, where approximations such as maximum-entropy or Gaussian phase packets (GPP) evolve the atomistic ensemble in a time-coarsened fashion. In practice, thi

P. Ghosh, J. Guo, F. Ye, T. Heitmann

Envisaging antiferromagnetic spintronics pivots on two key criteria of high transition temperature and tuning of underlying magnetic order using straightforward application of magnetic field or electric current. Here, we show that NiSi metal can provide suitable new platform in this quest. First, our study unveils high temperature antiferromagnetism in singl

Jure Kališnik

Action Lie groupoids are used to model spaces of orbits of actions of Lie groups on manifolds. For each such action groupoid $M\rtimes H$ we construct a locally convex bialgebroid $\mathord{\mathrm{Dirac}}(M\rtimes H)$ with an antipode over $\mathord{\mathcal{C}^{\infty}_{c}}(M)$, from which the groupoid $M\rtimes H$ can be reconstructed as its spectral acti

Michail Athanasakis-Kaklamanakis, Jordan R. Reilly, Agota Koszorus, Shane G. Wilkins

To optimize the performance of the Collinear Resonance Ionization Spectroscopy (CRIS) experiment at CERN-ISOLDE, technical upgrades are continuously introduced, aiming to enhance its sensitivity, precision, stability, and efficiency. Recently, a voltage-scanning setup was developed and commissioned at CRIS, which improved the scanning speed by a factor of th

Anil Alan, Tamas G. Molnar, Aaron D. Ames, Gábor Orosz

Deploying safety-critical controllers in practice necessitates the ability to modulate uncertainties in control systems. In this context, robust control barrier functions -- in a variety of forms -- have been used to obtain safety guarantees for uncertain systems. Yet the differing types of uncertainty experienced in practice have resulted in a fractured lan

Elvis do A. Soares, Amaro G. Barreto, Frederico W. Tavares

This study employs classical Density Functional Theory (cDFT) to investigate the adsorption isotherms and structural information of H2 and CH4 fluids inside MOF-5. The results indicate that the adsorption of both fluids is highly dependent on the fluid temperature and the shape of the MOF-5 structure. Specifically, the CH4 molecules exhibit stronger interact

Takanobu Aoyama

A vector topology on a vector space over a topological field is a (not necessarily Hausdorff) topology by which the addition and scalar multiplication are continuous. We prove that, if an isomorphism between the lattice of topologies of two vector spaces preserves vector topologies, then the isomorphism is induced by a translation, a semilinear isomorphism a

Antonio F. Rotundo, René Schwonnek

The entropic uncertainty principle in the form proven by Maassen and Uffink yields a fundamental inequality that is prominently used in many places all over the field of quantum information theory. In this work, we provide a family of versatile generalizations of this relation. Our proof methods build on a deep connection between entropic uncertainties and i

Zhengyuan Yang, Linjie Li, Jianfeng Wang, Kevin Lin

We propose MM-REACT, a system paradigm that integrates ChatGPT with a pool of vision experts to achieve multimodal reasoning and action. In this paper, we define and explore a comprehensive list of advanced vision tasks that are intriguing to solve, but may exceed the capabilities of existing vision and vision-language models. To achieve such advanced visual

Ik Jae Lee, David N Yetter

We recall Petit's construction of "dichromatic" invariants of 4-manifolds computed from Kirby diagrams using a nested pair of ribbon fusion categories $ B \subset C $ as initial data. Along the way we prove a lemma that fits the use of formal linear combinations of simple objects with quantum dimensions a coefficients as in the constructions of Reshetikhin-T

Joseph Hart, Mamikon Gulian, Indu Manickam, Laura Swiler

In complex large-scale systems such as climate, important effects are caused by a combination of confounding processes that are not fully observable. The identification of sources from observations of system state is vital for attribution and prediction, which inform critical policy decisions. The difficulty of these types of inverse problems lies in the ina

Zach Eidex, Yifu Ding, Jing Wang, Elham Abouei

MRI-guided radiation therapy (MRgRT) offers a precise and adaptive approach to treatment planning. Deep learning applications which augment the capabilities of MRgRT are systematically reviewed. MRI-guided radiation therapy offers a precise, adaptive approach to treatment planning. Deep learning applications which augment the capabilities of MRgRT are system

Jack G. Nedell, Michael Vogl, Gregory A. Fiete

Motivated by the observation and prediction of clustering behavior for impurities substituted into the host lattice of a real material, and the dramatic impact this can have on electronic properties, we develop a simple approach to describe such an effect via the electron self-energy. We employ a disorder averaged T-matrix expansion taken to second order, wh

Wenqi Wei, Mu Qiao, Divyesh Jadav

Graph Neural Networks (GNNs) have enjoyed wide spread applications in graph-structured data. However, existing graph based applications commonly lack annotated data. GNNs are required to learn latent patterns from a limited amount of training data to perform inferences on a vast amount of test data. The increased complexity of GNNs, as well as a single point

Daniel Bultrini, Oriol Vendrell

Mixed quantum-classical dynamics is a set of methods often used to understand systems too complex to treat fully quantum mechanically. Many techniques exist for full quantum mechanical evolution on quantum computers, but mixed quantum-classical dynamics are less explored. We present a modular algorithm for general mixed quantum-classical dynamics where the q

Alfred Gautschy

Dynamical stellar-evolution modeling through the AGB phase reveals that radial pulsations with very fast-growing amplitudes develop if the luminosity to mass ratio of stars with tenuous envelopes exceeds a critical limit. An instability going nonlinear already after a few pulsation cycles might qualify as a source of the superwind - postulated to shed a subs

Michael Chang, Alyssa L. Dayan, Franziska Meier, Thomas L. Griffiths

Object rearrangement is a challenge for embodied agents because solving these tasks requires generalizing across a combinatorially large set of configurations of entities and their locations. Worse, the representations of these entities are unknown and must be inferred from sensory percepts. We present a hierarchical abstraction approach to uncover these und

Alex Guo, Daniel Floryan, Michael D. Graham

This study aims to extract and characterize structures in fully developed pipe flow at a friction Reynolds number of $\text{Re}_\tau = 12\,400$. To do so, we employ data-driven wavelet decomposition (DDWD) [D.~Floryan and M.~D.~Graham, PNAS 118, e2021299118 (2021)], a method that combines features of proper orthogonal decomposition and wavelet analysis in or

Yinghao Wang, Rémi Nahon, Enzo Tartaglione, Pavlo Mozharovskyi

In this paper, we present a new approach to mental state classification from EEG signals by combining signal processing techniques and machine learning (ML) algorithms. We evaluate the performance of the proposed method on a dataset of EEG recordings collected during a cognitive load task and compared it to other state-of-the-art methods. The results show th

Thiago Carvalho Corso, Gero Friesecke

We derive the next order correction to the Dirac exchange energy for the free electron gas in a box with zero boundary conditions in the thermodynamic limit. The correction is of the order of the surface area of the box, and comes from three different contributions: (i) a real-space boundary layer, (ii) a boundary-condition-induced small shift of Fermi momen

Botao Hao, Rahul Jain, Dengwang Tang, Zheng Wen

In this paper, we address the following problem: Given an offline demonstration dataset from an imperfect expert, what is the best way to leverage it to bootstrap online learning performance in MDPs. We first propose an Informed Posterior Sampling-based RL (iPSRL) algorithm that uses the offline dataset, and information about the expert's behavioral policy u

Christopher Boettner, Maxime Trebitsch, Pratika Dayal

We study the co-evolution of dark matter halos, galaxies and supermassive black holes using an empirical galaxy evolution model from $z=0$ -- $10$. We demonstrate that by connecting dark matter structure evolution with simple empirical prescriptions for baryonic processes, we are able to faithfully reproduce key observations in the relation between galaxies

Vahagn Mkhitaryan, Andrew P. Weber, Saad Abdullah, Laura Fernández

The ability to confine light down to atomic scales is critical for the development of applications in optoelectronics and optical sensing as well as for the exploration of nanoscale quantum phenomena. Plasmons in metallic nanostructures can achieve this type of confinement, although fabrication imperfections down to the subnanometer scale hinder actual devel

Noah Shinn, Federico Cassano, Edward Berman, Ashwin Gopinath

Large language models (LLMs) have been increasingly used to interact with external environments (e.g., games, compilers, APIs) as goal-driven agents. However, it remains challenging for these language agents to quickly and efficiently learn from trial-and-error as traditional reinforcement learning methods require extensive training samples and expensive mod

Tomohiro Hirano, Alexis Akira Toda

We analyze how equilibrium housing prices are determined in the process of economic development within an overlapping generations model with perfect housing and rental markets. We characterize the rent growth rate in all equilibria. The economy exhibits a two-stage phase transition: as incomes of home buyers rise, the equilibrium regime changes from fundamen

Hareesh Gautham Bhaskar, Gongjie Li, Doug Lin

Recent studies have shown that AGN discs can host sources of gravitational waves. Compact binaries can form and merge in AGN discs through their interactions with the gas and other compact objects in the disc. It is also possible for the binaries to shorten the merging timescale due to eccentricity excitation caused by perturbations from the supermassive bla

Wei-Ting Chen, Wang Yifan, Sy-Yen Kuo, Gordon Wetzstein

Neural radiance fields (NeRFs) have demonstrated state-of-the-art performance for 3D computer vision tasks, including novel view synthesis and 3D shape reconstruction. However, these methods fail in adverse weather conditions. To address this challenge, we introduce DehazeNeRF as a framework that robustly operates in hazy conditions. DehazeNeRF extends the v

Adil Jueid, Jochem Kip, Roberto Ruiz de Austri, Peter Skands

In dark-matter annihilation channels to hadronic final states, stable particles -- such as positrons, photons, antiprotons, and antineutrinos -- are produced via complex sequences of phenomena including QED/QCD radiation, hadronisation, and hadron decays. These processes are normally modelled by Monte Carlo event generators whose limited accuracy imply intri

Nessim Sibony, Andrey Soldatenkov, Misha Verbitsky

A rigid cohomology class on a complex manifold is a class that is represented by a unique closed positive current. The positive current representing a rigid class is also called rigid. For a compact Kahler manifold $X$ all eigenvectors of hyperbolic automorphisms acting on $H^{1,1}(X)$ that have non-unit eigenvalues are rigid classes. Such classes are always

A. C. S. Readhead, V. Ravi, R. D. Blandford, A. G. Sullivan

We use a sample of 54 Compact Symmetric Objects (CSOs) to confirm that there are two unrelated CSO classes: an edge-dimmed, low-luminosity class (CSO~1), and an edge-brightened, high-luminosity class (CSO~2). Using blind tests, we show that CSO~2s consist of three sub-classes: CSO 2.0, having prominent hot-spots at the leading edges of narrow jets and/or nar

Sébastien Comerón, Ignacio Trujillo, Michele Cappellari, Fernando Buitrago

According to the $\Lambda$CDM cosmology, present-day galaxies with stellar masses $M_\star>10^{11} {\rm M}_\odot$ should contain a sizable fraction of dark matter within their stellar body. Models indicate that in massive early-type galaxies (ETGs) dark matter should account for $\sim60\%$ of the dynamical mass within five effective radii ($5 R_{\rm e}$). Mo

S. Kiehlmann, A. C. S. Readhead, S. O'Neill, P. N. Wilkinson

Compact Symmetric Objects (CSOs) are compact (<1 kpc), jetted Active Galactic Nuclei (AGN), whose jet axes are not aligned close to the line of sight, and whose observed emission is not predominantly relativistically boosted towards us. Two classes of CSOs have previously been identified: approximately one fifth are edge-dimmed and designated as CSO 1s, whil

D. W. Kurtz, R. Jayaraman, P. Sowicka, G. Handler

Several studies have shown that a number of stars pulsating in p modes lie between the $\beta$ Cep and $\delta$ Sct instability strips in the Hertzsprung-Russell (HR) Diagram. At present, there is no certain understanding of how p~modes can be excited in this $T_{\rm eff}$ range. The goal of this work is to disprove the conjecture that all stars pulsating in

S. Kiehlmann, M. L. Lister, A. C. S. Readhead, I. Liodakis

Compact Symmetric Objects (CSOs) are jetted Active Galactic Nuclei (AGN) with overall projected size <1 kpc. The classification was introduced to distinguish these objects from the majority of compact jetted-AGN in centimeter wavelength very long baseline interferometry observations, where the observed emission is relativistically boosted towards the observe

Eemeli Annala, Tyler Gorda, Joonas Hirvonen, Oleg Komoltsev

Neutron-star cores contain matter at the highest densities in our Universe. This highly compressed matter may undergo a phase transition where nuclear matter melts into deconfined quark matter, liberating its constituent quarks and gluons. Quark matter exhibits an approximate conformal symmetry, predicting a specific form for its equation of state (EoS), but

Roeland Wiersema, Dylan Lewis, David Wierichs, Juan Carrasquilla

Variational quantum algorithms use non-convex optimization methods to find the optimal parameters for a parametrized quantum circuit in order to solve a computational problem. The choice of the circuit ansatz, which consists of parameterized gates, is crucial to the success of these algorithms. Here, we propose a gate which fully parameterizes the special un

Mitchell J. Weikert, Matthew R. Buckley

Well-motivated models of dark matter often result in a population of electrons and positrons within galaxies produced through dark matter annihilation -- usually in association with gamma rays. As they diffuse through galactic magnetic fields, these $e^\pm$ produce synchrotron radio emission. The intensity and morphology of this signal depends on the propert

Ioannis D. Gialamas, Kyriakos Tamvakis

Bimetric gravity, is a theory of gravity that posits the existence of two interacting and dynamical metric tensors. The spectrum of bimetric gravity consists of a massless and a massive spin-2 particle. The form of the interactions between the two metrics $g_{\mu\nu}$ and $f_{\mu\nu}$ is constrained by requiring absence of the so called Boulware-Deser ghost.

Robert Heighton, Lucien Heurtier, Michael Spannowsky

Next-generation large-volume detectors, such as GRAND, POEMMA, Trinity, TAROGE-M, TAMBO, or PUEO, have been designed to search for ultra-high-energy cosmic rays (UHECRs) with unprecedented sensitivity. We propose to use these detectors to search for new physics beyond the Standard Model (BSM). By considering the simple case of a right-handed neutrino that mi

Sumit Banik, Andreas Crivellin, Syuhei Iguro, Teppei Kitahara

The two-Higgs-doublet model with a $U(1)_H$ gauge symmetry (N2HDM-$U(1)$) has several advantages compared to the ``standard'' $Z_2$ version (N2HDM-$Z_2$): It is purely based on gauge symmetries, involves only spontaneous symmetry breaking, and is more predictive because it contains one parameter less in the Higgs potential, which further ensures $CP$ conserv

Onur Günlü, Matthieu R. Bloch, Rafael F. Schaefer, Aylin Yener

This work considers the problem of mitigating information leakage between communication and sensing in systems jointly performing both operations. Specifically, a discrete memoryless state-dependent broadcast channel model is studied in which (i) the presence of feedback enables a transmitter to convey information, while simultaneously performing channel sta

Victoria Strait, Gabriel Brammer, Adam Muzzin, Guillaume Dezprez

We report the discovery of a low-mass $z=5.200\pm 0.002$ galaxy that is in the process of ceasing its star formation. The galaxy, MACS0417-z5PSB, is multiply imaged with magnification factors $\sim40$ by the galaxy cluster MACS J0417.5-1154, observed as part of the CAnadian NIRISS Unbiased Cluster Survey (CANUCS). Using observations of MACS0417-z5PSB with a

Daniel R. Weisz, Alessandro Savino, Andrew E. Dolphin

Using color-magnitude diagrams from deep archival Hubble Space Telescope imaging, we self-consistently measure the star formation history of Eridanus II (Eri II), the lowest-mass galaxy ($M_{\star}(z=0) \sim 10^5 M_{\odot}$) known to host a globular cluster (GC), and the age, mass, and metallicity of its GC. The GC ($\sim13.2\pm0.4$ Gyr, $\langle$[Fe/H]$\ran

Shih-Yun Tang, Asa G. Stahl, L. Prato, G. H. Schaefer

The stellar companion to the weak-line T Tauri star DI Tau A was first discovered by the lunar occultation technique in 1989 and was subsequently confirmed by a speckle imaging observation in 1991. It has not been detected since, despite being targeted by five different studies that used a variety of methods and spanned more than 20 years. Here, we report th

Matteo Robbiati, Juan M. Cruz-Martinez, Stefano Carrazza

The two main approaches to quantum computing are gate-based computation and analog computation, which are polynomially equivalent in terms of complexity, and they are often seen as alternatives to each other. In this work, we present a method for fitting one-dimensional probability distributions as a practical example of how analog and gate-based computation

Ammar Jahin, Yuxuan Wang

Superconductivity has been experimentally observed in monolayer WTe2, which in-plane field measurements suggested are of spin-triplet nature. Furthermore, it has been proposed that with a $p$-wave pairing, the material is a second-order topological superconductor with a pair of Majorana zero modes at the corners of a finite sample. We show that for a repulsi

Gordan Krnjaic, Tanner Trickle

Massive vector particles are minimal dark matter candidates that motivate a wide range of laboratory searches, primarily exploiting a postulated kinetic mixing with the photon. However, depending on the high energy field content, the dominant vector dark matter (VDM) coupling to visible particles may arise at higher operator dimension, motivating efforts to

Philipp Uhrich, Soumik Bandyopadhyay, Nick Sauerwein, Julian Sonner

The search for a quantum theory of gravity has led to the discovery of quantum many-body systems that are dual to gravitational models with quantum properties. The perhaps most famous of these systems is the Sachdev-Ye-Kitaev (SYK) model. It features maximal scrambling of quantum information, and opens a potential inroad to experimentally investigating aspec

Yuxin Fang, Quan Sun, Xinggang Wang, Tiejun Huang

We launch EVA-02, a next-generation Transformer-based visual representation pre-trained to reconstruct strong and robust language-aligned vision features via masked image modeling. With an updated plain Transformer architecture as well as extensive pre-training from an open & accessible giant CLIP vision encoder, EVA-02 demonstrates superior performance comp

Xuxin Cheng, Ashish Kumar, Deepak Pathak

Locomotion has seen dramatic progress for walking or running across challenging terrains. However, robotic quadrupeds are still far behind their biological counterparts, such as dogs, which display a variety of agile skills and can use the legs beyond locomotion to perform several basic manipulation tasks like interacting with objects and climbing. In this p

Ziyang Chen, Shengyi Qian, Andrew Owens

The images and sounds that we perceive undergo subtle but geometrically consistent changes as we rotate our heads. In this paper, we use these cues to solve a problem we call Sound Localization from Motion (SLfM): jointly estimating camera rotation and localizing sound sources. We learn to solve these tasks solely through self-supervision. A visual model pre

Ruoshi Liu, Rundi Wu, Basile Van Hoorick, Pavel Tokmakov

We introduce Zero-1-to-3, a framework for changing the camera viewpoint of an object given just a single RGB image. To perform novel view synthesis in this under-constrained setting, we capitalize on the geometric priors that large-scale diffusion models learn about natural images. Our conditional diffusion model uses a synthetic dataset to learn controls of

Yining Hong, Chunru Lin, Yilun Du, Zhenfang Chen

Humans are able to accurately reason in 3D by gathering multi-view observations of the surrounding world. Inspired by this insight, we introduce a new large-scale benchmark for 3D multi-view visual question answering (3DMV-VQA). This dataset is collected by an embodied agent actively moving and capturing RGB images in an environment using the Habitat simulat

Fabian Finger, Rodrigo Rosa-Medina, Nicola Reiter, Panagiotis Christodoulou

Engineering pairs of massive particles that are simultaneously correlated in their external and internal degrees of freedom is a major challenge, yet essential for advancing fundamental tests of physics and quantum technologies. In this Letter, we experimentally demonstrate a mechanism for generating pairs of atoms in well-defined spin and momentum modes. Th

Jihao Liu, Tai Wang, Boxiao Liu, Qihang Zhang

Multi-view camera-based 3D detection is a challenging problem in computer vision. Recent works leverage a pretrained LiDAR detection model to transfer knowledge to a camera-based student network. However, we argue that there is a major domain gap between the LiDAR BEV features and the camera-based BEV features, as they have different characteristics and are

Xi Chen, Shuang Li, Ser-Nam Lim, Antonio Torralba

Open-vocabulary image segmentation is attracting increasing attention due to its critical applications in the real world. Traditional closed-vocabulary segmentation methods are not able to characterize novel objects, whereas several recent open-vocabulary attempts obtain unsatisfactory results, i.e., notable performance reduction on the closed vocabulary and

Claudio Battiloro, Zhiyang Wang, Hans Riess, Paolo Di Lorenzo

In this work we introduce a convolution operation over the tangent bundle of Riemann manifolds in terms of exponentials of the Connection Laplacian operator. We define tangent bundle filters and tangent bundle neural networks (TNNs) based on this convolution operation, which are novel continuous architectures operating on tangent bundle signals, i.e. vector

Tian Zhou, Leonardo Modesto

We hereby focus on the analytic geodesic extension of several regular rotating black holes (RRBHs) obtained throughout the Newman-Janis algorithm starting from some popular spherically symmetric regular black holes. It turns out that if the metric is not an even function of Boyer-Lindquist radial coordinate r, similarly to the Kerr spacetime, the metric has

L. Barrufet, P. A. Oesch, R. Bouwens, H. Inami

We present the first observational infrared luminosity function (IRLF) measurement in the Epoch of Reionization (EoR) based on a UV-selected galaxy sample with ALMA spectroscopic observations. Our analysis is based on the ALMA large program Reionization Era Bright Emission Line Survey (REBELS), which targets 42 galaxies at $\mathrm{z=6.4-7.7}$ with [CII] 158

Xi Chen, Yau Shing Jonathan Cheung, Ser-Nam Lim, Hengshuang Zhao

Interactive segmentation enables users to extract masks by providing simple annotations to indicate the target, such as boxes, clicks, or scribbles. Among these interaction formats, scribbles are the most flexible as they can be of arbitrary shapes and sizes. This enables scribbles to provide more indications of the target object. However, previous works mai

Yuding Liu, Dongzhu Liu, Guangxu Zhu, Qingjiang Shi

Over-the-air federated edge learning (Air-FEEL) is a communication-efficient framework for distributed machine learning using training data distributed at edge devices. This framework enables all edge devices to transmit model updates simultaneously over the entire available bandwidth, allowing for over-the-air aggregation. A one-bit digital over-the-air agg

Adrien Houge, Sebastiaan Krijt

Most protoplanetary discs are thought to undergo violent and frequent accretion outbursts, during which the accretion rate and central luminosity are elevated for several decades. This temporarily increases the disc temperature, leading to the sublimation of ice species as snowlines move outwards. In this paper, we investigate how an FUor-type accretion outb

E. M. Stoudenmire, Xavier Waintal

Grover's algorithm is a primary algorithm offered as evidence that quantum computers can provide an advantage over classical computers. It involves an "oracle" specified for a given application whose structure is not part of the formal scaling of the quadratic speedup guaranteed by the algorithm. Grover's algorithm also requires exponentially many calls to t

Jiaqi Chen, Jiachen Lu, Xiatian Zhu, Li Zhang

We present Generative Semantic Segmentation (GSS), a generative learning approach for semantic segmentation. Uniquely, we cast semantic segmentation as an image-conditioned mask generation problem. This is achieved by replacing the conventional per-pixel discriminative learning with a latent prior learning process. Specifically, we model the variational post

Wenxuan Zhou, Sheng Zhang, Hoifung Poon, Muhao Chen

Large language models (LLMs) encode parametric knowledge about world facts and have shown remarkable performance in knowledge-driven NLP tasks. However, their reliance on parametric knowledge may cause them to overlook contextual cues, leading to incorrect predictions in context-sensitive NLP tasks (e.g., knowledge acquisition tasks). In this paper, we seek

Harry Desmond

The radial acceleration relation (RAR) of late-type galaxies relates their dynamical acceleration, $g_\text{obs}$, to that sourced by baryons alone, $g_\text{bar}$, across their rotation curves. Literature fits to the RAR have fixed the galaxy parameters on which the relation depends -- distance, inclination, luminosity and mass-to-light ratios -- to their m

B. Hamil, B. C. Lütfüoğlu

In the framework of the theory of Dunkl-deformed bosons, Bose-Einstein condensation of two and three-dimensional Dunkl-boson gases confined in the one-dimensional gravitational field is investigated. Using the semi-classical approximation method, we calculate the expressions of the Dunkl-critical temperature $T_{c}^{D}$, the ground state population $\frac{N_

Deepti Hegde, Jeya Maria Jose Valanarasu, Vishal M. Patel

Vision-Language models like CLIP have been widely adopted for various tasks due to their impressive zero-shot capabilities. However, CLIP is not suitable for extracting 3D geometric features as it was trained on only images and text by natural language supervision. We work on addressing this limitation and propose a new framework termed CG3D (CLIP Goes 3D) w

Michael J Mahoney

Assessing predictive models can be challenging. Modelers must navigate a wide array of evaluation methodologies implemented with incompatible interfaces across multiple packages which may give different or even contradictory results, while ensuring that their chosen approach properly estimates the performance of their model when generalizing to new observati

D. A. Baiko

Based on the linear mixing approach, we calculate the latent heat for crystallizing fully-ionized $^{12}$C/$^{16}$O and $^{16}$O/$^{20}$Ne mixtures in white dwarf (WD) cores for two different parametrizations of the corrections to the linear-mixing energies and with account of ion quantum effects. We report noticeable composition-dependent deviations of the

Priyanka Kaswan, Sennur Ulukus

We consider a semantics-aware communication system, where timeliness is the semantic measure, with a source which maintains the most current version of a file, and a network of $n$ user nodes with the goal to acquire the latest version of the file. The source gets updated with newer file versions as a point process, and forwards them to the user nodes, which

Song Sun

We prove the finite step termination of bubble trees for singularity formation of polarized K\"ahler-Einstein metrics in the non-collapsing situation. We also raise several questions and conjectures in connection with algebraic geometry and Riemannian geometry.

Franziska A. Breitner, Anton Jesche, Vladimir Tsurkan, Philipp Gegenwart

We explore the effect of heat treatment in argon atmosphere under various temperatures up to $500^\circ$C on single crystals of $\alpha$-RuCl$_3$ by study of the mass loss, microprobe energy dispersive x-ray spectroscopy, powder x-ray diffraction, electrical resistance as well as low-temperature magnetic susceptibility and specific heat. Clear signatures of

Shu-Hao Yeh, Shuangyu Xie, Di Wang, Wei Yan

Optical Image Stabilization (OIS) system in mobile devices reduces image blurring by steering lens to compensate for hand jitters. However, OIS changes intrinsic camera parameters (i.e. $\mathrm{K}$ matrix) dynamically which hinders accurate camera pose estimation or 3D reconstruction. Here we propose a novel neural network-based approach that estimates $\ma

Or Patashnik, Daniel Garibi, Idan Azuri, Hadar Averbuch-Elor

Text-to-image models give rise to workflows which often begin with an exploration step, where users sift through a large collection of generated images. The global nature of the text-to-image generation process prevents users from narrowing their exploration to a particular object in the image. In this paper, we present a technique to generate a collection o

Ligong Han, Yinxiao Li, Han Zhang, Peyman Milanfar

Diffusion models have achieved remarkable success in text-to-image generation, enabling the creation of high-quality images from text prompts or other modalities. However, existing methods for customizing these models are limited by handling multiple personalized subjects and the risk of overfitting. Moreover, their large number of parameters is inefficient

Roy Araiza, Yidong Chen, Marius Junge, Peixue Wu

We introduce a new framework for quantifying the complexity of quantum channels, grounded in a suitably chosen resource set. This class of convex functions is designed to analyze the complexity of both open and closed quantum systems. By leveraging Lipschitz norms inspired by quantum optimal transport theory, we rigorously establish the fundamental propertie

Johannes Greil, Matthias Golibrzuch, Martina Kiechle, Ádám Papp

This work examines the impact of electromagnetic cross-talk in magnonic devices when using inductive spin-wave (SW) transducers. We present detailed electrical SW spectroscopy measurements showing the signal contributions to be considered in magnonic device design. We further provide a rule of thumb estimation for the cross-talk that is responsible for the s

Weixuan Sun, Jiayi Zhang, Jianyuan Wang, Zheyuan Liu

Self-supervised audio-visual source localization aims to locate sound-source objects in video frames without extra annotations. Recent methods often approach this goal with the help of contrastive learning, which assumes only the audio and visual contents from the same video are positive samples for each other. However, this assumption would suffer from fals

Yukang Chen, Jianhui Liu, Xiangyu Zhang, Xiaojuan Qi

3D object detectors usually rely on hand-crafted proxies, e.g., anchors or centers, and translate well-studied 2D frameworks to 3D. Thus, sparse voxel features need to be densified and processed by dense prediction heads, which inevitably costs extra computation. In this paper, we instead propose VoxelNext for fully sparse 3D object detection. Our core insig

Beata Gryszka, Janusz Gwoździewicz

Let $x=t^n$, $y=\sum_{i=1}^{\infty}a_it^i$ be a parametrisation of the germ of a complex plane analytic curve $\Gamma$ at the origin. Then $\Gamma$ has the implicit equation $f(x,y)=0$ in the neighbourhood of the origin, where $f=\sum c_{ij}x^iy^j$ is a Weierstrass polynomial in $\mathbb{C}[[x]][y]$ of degree $n$. Every polynomial depending on coefficients $

T. Otsuka, Y. Tsunoda, N. Shimizu, Y. Utsuno

In the traditional view, heavy deformed nuclei are like axially-symmetric prolate ellipsoids, rotating about one of the short axes. In the present picture, their shapes may be triaxial. The triaxial shape yields complex rotations, which actually well reproduce experimental data, as confirmed by state-of-the-art Configuration Interaction calculations. Two ori

Pau de Jorge, Riccardo Volpi, Philip Torr, Gregory Rogez

Motivated by the increasing popularity of transformers in computer vision, in recent times there has been a rapid development of novel architectures. While in-domain performance follows a constant, upward trend, properties like robustness or uncertainty estimation are less explored -leaving doubts about advances in model reliability. Studies along these axes

Jacek Jendrej, Andrew Lawrie

We consider a real scalar field equation in dimension 1+1 with an even positive self-interaction potential having two non-degenerate zeros (vacua) 1 and -1. It is known that such a model admits non-trivial static solutions called kinks and antikinks. A kink cluster is a solution approaching, for large positive times, a superposition of alternating kinks and