May 2023 arXiv papers — page 60

Mark Van der Merwe, Youngsun Wi, Dmitry Berenson, Nima Fazeli

Reasoning over the interplay between object deformation and force transmission through contact is central to the manipulation of compliant objects. In this paper, we propose Neural Deforming Contact Field (NDCF), a representation that jointly models object deformations and contact patches from visuo-tactile feedback using implicit representations. Representi

Ye Mou, Xingyu Yang, Bruno Gallas, Mathieu Mivelle

The inverse Faraday effect is a magneto-optical process allowing the magnetization of matter by an optical excitation carrying a non-zero spin of light. In particular, a right circular polarization generates a magnetization in the direction of light propagation and a left circular polarization in the opposite direction to this propagation. We demonstrate her

Sophia Harrison, Eleonora Gualdoni, Gemma Boleda

Gender bias in Language and Vision datasets and models has the potential to perpetuate harmful stereotypes and discrimination. We analyze gender bias in two Language and Vision datasets. Consistent with prior work, we find that both datasets underrepresent women, which promotes their invisibilization. Moreover, we hypothesize and find that a bias affects hum

Anatol Garioud, Apolline De Wit, Marc Poupée, Marion Valette

The FLAIR #2 dataset hereby presented includes two very distinct types of data, which are exploited for a semantic segmentation task aimed at mapping land cover. The data fusion workflow proposes the exploitation of the fine spatial and textural information of very high spatial resolution (VHR) mono-temporal aerial imagery and the temporal and spectral richn

Soumik Bhattacharya, Vandana Tripathi, S. L. Tabor, A. Volya

Results from the study of $\beta^-$-decay of $^{45}$Cl, produced in the fragmentation of a 140-MeV/u $^{48}$Ca beam, are presented. The half-life for $^{45}$Cl $\beta$-decay is measured to be 513(36) ms. The $\beta^-$ and $\beta^- 1n$ decay of $^{45}$Cl populated excited states in $^{45,44}$Ar, respectively. On the basis of $\gamma$-ray singles and $\gamma$-

Chao Li, Michael Rapoport, Wei Zhang

We define Hecke correspondences and Hecke operators on unitary RZ spaces and study their basic geometric properties, including a commutativity conjecture on Hecke operators. Then we formulate the Arithmetic Fundamental Lemma conjecture for the spherical Hecke algebra. We also formulate a conjecture on the abundance of spherical Hecke functions with identical

Doyeol Ahn

In this study, we explore the non-Markovian cost function for quantum error mitigation (QEM) and the representation of two-qubit operators using Dirac Gamma matrices, central to the structure of relativistic quantum mechanics. The primary focus of quantum computing research, particularly with noisy intermediate-scale quantum (NISQ) devices, is on reducing er

Tarek Naous, Michael J. Ryan, Anton Lavrouk, Mohit Chandra

We present a comprehensive evaluation of large language models for multilingual readability assessment. Existing evaluation resources lack domain and language diversity, limiting the ability for cross-domain and cross-lingual analyses. This paper introduces ReadMe++, a multilingual multi-domain dataset with human annotations of 9757 sentences in Arabic, Engl

Hanlin Mo, Guoying Zhao

The topic of achieving rotational invariance in convolutional neural networks (CNNs) has gained considerable attention recently, as this invariance is crucial for many computer vision tasks such as image classification and matching. In this letter, we propose a Sorting Convolution (SC) inspired by some hand-crafted features of texture images, which achieves

Diego Arroyuelo, Gabriel Carmona, Héctor Larrañaga, Francisco Riveros

Large-alphabet strings are common in scenarios such as information retrieval and natural-language processing. The efficient storage and processing of such strings usually introduces several challenges that are not witnessed in small-alphabets strings. This paper studies the efficient implementation of one of the most effective approaches for dealing with lar

Mohsen Ahmadi, Ahmad Gholizadeh Lonbar, Mohammadsadegh Nouri, Amir Sharifzadeh Javidi

This study explores the use of a digital twin model and deep learning method to build a global terrain and altitude map based on USGS information. The goal is to artistically represent various landforms while incorporating precise elevation modifications in the terrain map and encoding land height in the altitude map. A random selection of 5000 segments from

Yunzhe Li, Qian Chen, Weixiang Yan, Wen Wang

Existing works on outline-conditioned text generation typically aim to generate text using provided outlines as rough sketches, such as keywords and phrases. However, these approaches make it challenging to control the quality of text generation and assess consistency between outlines and generated texts due to lack of clarity and rationality of the rough ou

David Heineman, Yao Dou, Mounica Maddela, Wei Xu

Large language models (e.g., GPT-4) are uniquely capable of producing highly rated text simplification, yet current human evaluation methods fail to provide a clear understanding of systems' specific strengths and weaknesses. To address this limitation, we introduce SALSA, an edit-based human annotation framework that enables holistic and fine-grained text s

Maha Tufail Agro, Hanan Aldarmaki

Labels noise refers to errors in training labels caused by cheap data annotation methods, such as web scraping or crowd-sourcing, which can be detrimental to the performance of supervised classifiers. Several methods have been proposed to counteract the effect of random label noise in supervised classification, and some studies have shown that BERT is alread

Mengxia Yu, Zhihan Zhang, Wenhao Yu, Meng Jiang

Comparative reasoning is a process of comparing objects, concepts, or entities to draw conclusions, which constitutes a fundamental cognitive ability. In this paper, we propose a novel framework to pre-train language models for enhancing their abilities of comparative reasoning over texts. While there have been approaches for NLP tasks that require comparati

Tarek Naous, Michael J. Ryan, Alan Ritter, Wei Xu

As the reach of large language models (LMs) expands globally, their ability to cater to diverse cultural contexts becomes crucial. Despite advancements in multilingual capabilities, models are not designed with appropriate cultural nuances. In this paper, we show that multilingual and Arabic monolingual LMs exhibit bias towards entities associated with Weste

Bruno L. Canto Martins, Yuri S. Messias, Maria I. Arruda Gonçalves, Izan C. Leão

Star-planet interactions play, among other things, a crucial role in planetary orbital configurations by circularizing orbits, aligning the star and planet spin and synchronizing stellar rotation with orbital motions. This is especially true for innermost giant planets, which can be schematized as binary systems with a very large mass ratio. Despite a few ex

Sebastian Ober, Ben Anson, Edward Milsom, Laurence Aitchison

Deep kernel processes are a recently introduced class of deep Bayesian models that have the flexibility of neural networks, but work entirely with Gram matrices. They operate by alternately sampling a Gram matrix from a distribution over positive semi-definite matrices, and applying a deterministic transformation. When the distribution is chosen to be Wishar

Pavan Kalyan Reddy Neerudu, Subba Reddy Oota, Mounika Marreddy, Venkateswara Rao Kagita

Transformer-based pretrained models like BERT, GPT-2 and T5 have been finetuned for a large number of natural language processing (NLP) tasks, and have been shown to be very effective. However, while finetuning, what changes across layers in these models with respect to pretrained checkpoints is under-studied. Further, how robust are these models to perturba

Qidong Yang, Alex Hernandez-Garcia, Paula Harder, Venkatesh Ramesh

Climate simulations are essential in guiding our understanding of climate change and responding to its effects. However, it is computationally expensive to resolve complex climate processes at high spatial resolution. As one way to speed up climate simulations, neural networks have been used to downscale climate variables from fast-running low-resolution sim

Mohit Yadav, Daniel Sheldon, Cameron Musco

Structured kernel interpolation (SKI) accelerates Gaussian process (GP) inference by interpolating the kernel covariance function using a dense grid of inducing points, whose corresponding kernel matrix is highly structured and thus amenable to fast linear algebra. Unfortunately, SKI scales poorly in the dimension of the input points, since the dense grid si

Ridong Han, Chaohao Yang, Tao Peng, Prayag Tiwari

Human-like large language models (LLMs), especially the most powerful and popular ones in OpenAI's GPT family, have proven to be very helpful for many natural language processing (NLP) related tasks. Therefore, various attempts have been made to apply LLMs to information extraction (IE), which is a fundamental NLP task that involves extracting information fr

Zheng Chen, Ziyan Jiang, Fan Yang, Eunah Cho

Conversational AI systems such as Alexa need to understand defective queries to ensure robust conversational understanding and reduce user friction. These defective queries often arise from user ambiguities, mistakes, or errors in automatic speech recognition (ASR) and natural language understanding (NLU). Personalized query rewriting is an approach that foc

Daniel S. Graça, Ning Zhong

In this paper, we examine the relationship between the stability of the dynamical system $x^{\prime}=f(x)$ and the computability of its basins of attraction. We present a computable $C^{\infty}$ system $x^{\prime}=f(x)$ that possesses a computable and stable equilibrium point, yet whose basin of attraction is robustly non-computable in a neighborhood of $f$

Joanne L. Pledger, Michael M. Shara

Stellar evolution theory predicts multiple pathways to the explosive deaths of stars as supernovae. Locating and characterizing the progenitors of well-studied supernovae is important to constrain the theory, and to justify and design future surveys to improve on progenitor detections. Here we report the serendipitous pre-explosion imaging, by the {\it Hubbl

Marcos A. G. Garcia, Mathias Pierre, Sarunas Verner

Conventional scenarios of purely gravitationally produced dark matter with masses below the Hubble parameter at the end of inflation are in tension with Cosmic Microwave Background (CMB) constraints on the isocurvature power spectrum. We explore a more general scenario with a non-minimal coupling between the scalar dark matter field and gravity, which allows

Jenny Chen, Benjamin Ades-Aron, Hong-Hsi Lee, Subah Mehrin

Various diffusion MRI (dMRI) preprocessing pipelines are currently available to yield more accurate diffusion parameters. Here, we evaluated accuracy and robustness of the optimized Diffusion parameter EStImation with Gibbs and NoisE Removal (DESIGNER) pipeline in a large clinical dMRI dataset and using ground truth phantoms. DESIGNER has been modified to im

Valerio De Luca, Justin Khoury, Sam S. C. Wong

Tidal Love numbers describe the linear response of a compact object under the presence of external tidal perturbations, and they are found to vanish exactly for black holes within General Relativity. In this paper we investigate the tidal deformability of neutral black holes when non-linearities in the theory are taken into account. As a case in point, we co

J. Stanton, R. Sharankova, K. Seiya, M. Wesley

The Fermilab Linac is a roughly 145 meter linear accelerator that accelerates H- beam from 750 keV to 400 MeV and provides beam for the Booster and the rest of the accelerator chain. The first section of the Linac is a Drift-Tube Linac (DTL), which in its current state, suffers from a lack of instrumentation along its length. As a result, operational staff d

Michalis K. Titsias

We define an optimal preconditioning for the Langevin diffusion by analytically optimizing the expected squared jumped distance. This yields as the optimal preconditioning an inverse Fisher information covariance matrix, where the covariance matrix is computed as the outer product of log target gradients averaged under the target. We apply this result to the

Zhihan Zhang, Wenhao Yu, Zheng Ning, Mingxuan Ju

Contrast consistency, the ability of a model to make consistently correct predictions in the presence of perturbations, is an essential aspect in NLP. While studied in tasks such as sentiment analysis and reading comprehension, it remains unexplored in open-domain question answering (OpenQA) due to the difficulty of collecting perturbed questions that satisf

Andrea Placidi, Gianluca Grignani, Troels Harmark, Marta Orselli

We provide the post-Newtonian (PN) waveform for binary systems in motion along generic planar orbits at 2.5PN accuracy, in terms of the dynamical variables of the effective one-body (EOB) formalism. In addition to the calculation of the higher order terms for all the contributions to the waveform that have been already considered in previous avatars of EOB m

Ewelina Mulawa

In this paper we study QCH K\"ahler surfaces, i.e. 4-dimensional Riemannian manifolds (of signature (++++)) admitting a K\"ahler complex structure with quasi-constant holomorphic sectional curvature. We give a detailed description of QCH K\"ahler surfaces of generalized Calabi type.

Ammon Fischer, Lennart Klebl, Jonas B. Profe, Alexander Rothstein

Motivated by the recent experimental realization of ABCB stacked tetralayer graphene [Wirth et al., ACS Nano 16, 16617 (2022)], we study correlated phenomena in moir\'e-less graphene tetralayers for realistic interaction profiles using an orbital resolved random phase approximation approach. We demonstrate that magnetic fluctuations originating from local in

Mattia C. Sormani, Ashley T. Barnes, Jiayi Sun, Sophia K. Stuber

Galactic bars can drive cold gas inflows towards the centres of galaxies. The gas transport happens primarily through the so-called bar ``dust lanes'', which connect the galactic disc at kpc scales to the nuclear rings at hundreds of pc scales much like two gigantic galactic rivers. Once in the ring, the gas can fuel star formation activity, galactic outflow

Giles Blaney, Fernando Ivich, Angelo Sassaroli, Mark Niedre

Significance: Diffuse in-vivo Flow Cytometry (DiFC) is an emerging fluorescence sensing method to non-invasively detect labeled circulating cells in-vivo. However, due to Signal-to-Noise Ratio (SNR) constraints largely attributed to background tissue autofluorescence, DiFC's measurement depth is limited. multiplies Aim: The Dual-Ratio (DR) / dual-slope is a

Amir Sharon, Doron Kushnir, Wenlong Yuan, Lucas Macri

The SH0ES collaboration Hubble constant determination is in a ${\sim}5\sigma$ difference with the $Planck$ value, known as the Hubble tension. The accuracy of the Hubble constant measured with extragalactic Cepheids depends on robust stellar-crowding background estimation. Riess et al. (R20) compared the light curves amplitudes of extragalactic and MW Cephei

Yew Ken Chia, Hui Chen, Wei Han, Guizhen Chen

Aspect Sentiment Triplet Extraction (ASTE) is a challenging task in sentiment analysis, aiming to provide fine-grained insights into human sentiments. However, existing benchmarks are limited to two domains and do not evaluate model performance on unseen domains, raising concerns about the generalization of proposed methods. Furthermore, it remains unclear i

Daniel Parrochia

We first recall fundamentals of elementary climate physics: solar constant, radiative balance, greenhouse effect, astronomical parameters of the climate (theory of Milankovitch). Without disputing the analyzes of climatologists and the famous Keeling curve revealing in an indisputable way the increase in CO$_{2}$ in the atmosphere since the industrial revolu

Tsun Hin Navin Tsung, S. Peng Oh, Chad Bustard

We investigate how cosmic rays (CRs) affect thermal and hydrostatic stability of circumgalactic (CGM) gas, in simulations with both CR streaming and diffusion. Local thermal instability can be suppressed by CR-driven entropy mode propagation, in accordance with previous analytic work. However, there is only a narrow parameter regime where this operates, befo

Avia Noah, Yishay Zur, Nofar Fridman, Sourabh Singh

The synthesis of two-dimensional van der Waals magnets has paved the way for both technological applications and fundamental research on magnetism confined to ultra-small length scales. Edge magnetic moments in ferromagnets are expected to be less magnetized than in the sample interior because of the reduced amount of neighboring ferromagnetic spins at the s

P. G. Ortega, D. R. Entem, F. Fernandez, J. Segovia

Using proton-proton collisions at centre-of-mass energies $7$, $8$, and $13$ TeV, with a total integrated luminosity of $9\,\text{fb}^{-1}$, the LHCb collaboration has performed amplitude analyses of the $B^+\to D^+D^-K^+$, $B^+\to D^- D_s^+ \pi^+$ and $B^0\to \bar{D}^0 D_s^+ \pi^-$ decays, observing that new $T_{cs}$ and $T_{c\bar s}$ resonances are require

B. King

The collision of a finite electromagnetic plane wave with an electron subject to the Landau-Lifshitz radiation reaction force is studied. A locally monochromatic approximation is derived and compared to numerical evaluation of the exact plane wave result. Energy and transverse momentum spectra are calculated, which clearly display the red-shifting of harmoni

Wentao Bao, Lichang Chen, Heng Huang, Yu Kong

Compositional zero-shot learning (CZSL) task aims to recognize unseen compositional visual concepts, e.g., sliced tomatoes, where the model is learned only from the seen compositions, e.g., sliced potatoes and red tomatoes. Thanks to the prompt tuning on large pre-trained visual language models such as CLIP, recent literature shows impressively better CZSL p

S. Sandner, P. Hernandez, J. Lopez-Pavon, N. Rius

We consider the generation of a baryon asymmetry in an extension of the Standard Model with two singlet Majorana fermions that are degenerate above the electroweak phase transition. The model can explain neutrino masses as well as the observed matter-antimatter asymmetry, for masses of the heavy singlets below the electroweak scale. The only physical CP viol

Andrea Solfanelli, Stefano Ruffo, Sauro Succi, Nicolò Defenu

The simulation of complex quantum many-body systems is a promising short-term goal of noisy intermediate-scale quantum (NISQ) devices. However, the limited connectivity of native qubits hinders the implementation of quantum algorithms that require long-range interactions. We present the outcomes of a digital quantum simulation where we overcome the limitatio

Mohamed M. Anber, Erich Poppitz

We study $4$-dimensional $SU(N)\times U(1)$ gauge theories with a single massless Dirac fermion in the $2$-index symmetric/antisymmetric representations and show that they are endowed with a noninvertible $0$-form $\widetilde {\mathbb Z}_{2(N\pm 2)}^{\chi}$ chiral symmetry along with a $1$-form $\mathbb Z_N^{(1)}$ center symmetry. By using the Hamiltonian fo

Brent Tan, Drummond B. Fielding

Galaxies comprise intricate networks of interdependent processes which together govern their evolution. Central among these are the multiplicity of feedback channels, which remain incompletely understood. One outstanding problem is the understanding and modeling of the multiphase nature of galactic winds, which play a crucial role in galaxy formation and evo

Valentin Crépel, Aaron Dunbrack, Daniele Guerci, John Bonini

We demonstrate that a single layer of graphene subject to a superlattice potential nearly commensurate to a $\sqrt{3} \times \sqrt{3}$ supercell exactly maps to the chiral model of twisted bilayer graphene, albeit with half as many degrees of freedom. We comprehensively review the properties of this ``half-chiral model,'' including the interacting phases sta

Misha Yutushui, David F. Mross

Non-Abelian phases are among the most highly-sought states of matter, with those whose anyons permit universal quantum gates constituting the ultimate prize. The most promising candidate of such a phase is the fractional quantum Hall plateau at filling factors $\nu=\frac{12}{5}$, which putatively facilitates Fibonacci anyons. Experimental validation of this

Alex I. Malz, Mi Dai, Kara A. Ponder, Emille E. O. Ishida

Context: When selecting a classifier to use for a supernova Ia (SN Ia) cosmological analysis, it is common to make decisions based on metrics of classification performance, i.e. contamination within the photometrically classified SN Ia sample, rather than a measure of cosmological constraining power. If the former is an appropriate proxy for the latter, this

Francesco D'Eramo, Giuseppe Lucente, Newton Nath, Seokhoon Yun

Solar nuclear reactions can occasionally produce sub-MeV elusive beyond the Standard Model particles that escape the solar interior without further interactions. This study focuses on massive spin-one particles. We construct the general theoretical framework and identify two crucial mixing sources involving the photon, which facilitate communication between

Constant Auclair, Erwan Allys, François Boulanger, Matthieu Béthermin

The low brightness dust emission at high Galactic latitude is of interest to study the interplay between physical processes in shaping the structure of the interstellar medium (ISM), as well as to statistically characterize dust emission as a foreground to the Cosmic Microwave Background (CMB). Progress in this avenue of research have been hampered by the di

Guillermo Barro, Pablo G. Perez-Gonzalez, Dale D. Kocevski, Elizabeth J. McGrath

We study a new population of extremely red objects (EROs) recently discovered by JWST based on their NIRCam colors F277W$-$F444W $>1.5$ mag. We find 37 EROs in the CEERS field with F444W $<28$ mag and photometric redshifts between $5<z<7$, with median $z=6.9^{+1.0}_{-1.6}$. Surprisingly, despite their red long-wavelength colors, these EROs have blue short-wa

Daniil Kochergin, Ivan M. Khaymovich, Olga Valba, Alexander Gorsky

We consider the properties of the random regular graph with node degree $d$ perturbed by chemical potentials $\mu_k$ for a number of short $k$-cycles. We analyze both numerically and analytically the phase diagram of the model in the $(\mu_k,d)$ plane. The critical curve separating the homogeneous and clusterized phases is found and it is demonstrated that t

Alexandros Ziampras, Richard P. Nelson, Roman R. Rafikov

ALMA observations of protoplanetary disks in dust continuum emission reveal a variety of annular structures. Attributing the existence of such features to embedded planets is a popular scenario, supported by studies using hydrodynamical models. Recent work has shown that radiative cooling greatly influences the capability of planet-driven spiral density wave

Annalisa Citro, Danielle A. Berg, Dawn K. Erb, Matthew W. Auger

We provide one of the most comprehensive metallicity studies at z $\sim$ 4 by analyzing the UV/optical HST photometry, and rest-frame VLT-FORS2 ultraviolet and VLT-XSHOOTER optical spectra of J0332-3557, a gravitationally lensed galaxy magnified by a factor of 20. With a 5$\sigma$ detection of the auroral O III]1666 line, we are able to derive a direct gas m

E. Vanzella, F. Loiacono, P. Bergamini, U. Mestric

We present JWST/NIRSpec integral field spectroscopy (IFS) of a lensed Population III candidate stellar complex (dubbed Lensed And Pristine 1, LAP1), with a lensing-corrected stellar mass ~<10^4 Msun, absolute luminosity M_UV > -11.2 (m_UV > 35.6), confirmed at redshift 6.639 +/- 0.004. The system is strongly amplified (\mu >~ 100) by straddling a critical li

Pierre Heidmann, Nicholas Speeney, Emanuele Berti, Ibrahima Bah

We study scalar perturbations of topological solitons, smooth horizonless solutions in five-dimensional Einstein-Maxwell theory that correspond to coherent states of gravity via the dynamics of extra compact dimensions. First, we compute scalar quasinormal modes for topological stars that have a single unstable photon sphere, and we show that the spectrum is

Temple He, Ana-Maria Raclariu, Kathryn M. Zurek

We study the relationship between shockwave geometries and the gravitational memory effect in four-dimensional asymptotically flat spacetime. In particular, we show the 't Hooft commutation relations of shockwave operators are equivalent to the commutation relation between soft and Goldstone modes parametrizing a sector of the gravitational phase space. We d

Robert E. Throckmorton, S. Das Sarma

The work of Ahn derives the noise power spectrum of a two-level fluctuator (TLF) in the case that it interacts only with a subregion of a full electron bath and thus is subject to a fluctuating temperature. However, Eq.~(1), which gives the variance of the subbath temperature in terms of the heat capacity, in that work carries the implicit assumption that th

Nikhil Sarin, Axel Brandenburg, Brynmor Haskell

The origin and evolution of magnetic fields of neutron stars from birth has long been a source of debate. Here, motivated by recent simulations of the Hall cascade with magnetic helicity, we invoke a model where the large-scale magnetic field of neutron stars grows as a product of small-scale turbulence through an inverse cascade. We apply this model to a si

Theresa C. Anderson, Angel V. Kumchev, Eyvindur A. Palsson

We decompose the discrete bilinear spherical averaging operator into simpler operators in several ways. This leads to a wide array of extensions, such as to the simplex averaging operator, and applications, such as to operator bounds.

Taeksoo Kim, Shunsuke Saito, Hanbyul Joo

Deep generative models have been recently extended to synthesizing 3D digital humans. However, previous approaches treat clothed humans as a single chunk of geometry without considering the compositionality of clothing and accessories. As a result, individual items cannot be naturally composed into novel identities, leading to limited expressiveness and cont

Agrim Gupta, Jiajun Wu, Jia Deng, Li Fei-Fei

Establishing correspondence between images or scenes is a significant challenge in computer vision, especially given occlusions, viewpoint changes, and varying object appearances. In this paper, we present Siamese Masked Autoencoders (SiamMAE), a simple extension of Masked Autoencoders (MAE) for learning visual correspondence from videos. SiamMAE operates on

Alejandro Escontrela, Ademi Adeniji, Wilson Yan, Ajay Jain

Specifying reward signals that allow agents to learn complex behaviors is a long-standing challenge in reinforcement learning. A promising approach is to extract preferences for behaviors from unlabeled videos, which are widely available on the internet. We present Video Prediction Rewards (VIPER), an algorithm that leverages pretrained video prediction mode

Hong Liu, Zhiyuan Li, David Hall, Percy Liang

Given the massive cost of language model pre-training, a non-trivial improvement of the optimization algorithm would lead to a material reduction on the time and cost of training. Adam and its variants have been state-of-the-art for years, and more sophisticated second-order (Hessian-based) optimizers often incur too much per-step overhead. In this paper, we

Yue Guo, Tal August, Gondy Leroy, Trevor Cohen

While there has been significant development of models for Plain Language Summarization (PLS), evaluation remains a challenge. PLS lacks a dedicated assessment metric, and the suitability of text generation evaluation metrics is unclear due to the unique transformations involved (e.g., adding background explanations, removing jargon). To address these questi

R. Kailasham, Aditya S. Khair

Numerical simulations are performed on the collective dynamics of active disks, whose self-propulsion speed ($U$) varies in time, and whose orientation evolves according to rotational Brownian motion. Two protocols for the evolution of speed are considered: (i) a deterministic one involving a periodic change in $U$ at a frequency $\omega$; and (ii) a stochas

Bin Zhuang, Noé Lugaz, Nada Al-Haddad, Réka M. Winslow

A fundamental property of coronal mass ejections (CMEs) is their radial expansion, which determines the increase in the CME radial size and the decrease in the CME magnetic field strength as the CME propagates. CME radial expansion can be investigated either by using remote observations or by in-situ measurements based on multiple spacecraft in radial conjun

Harman Singh, Poorva Garg, Mohit Gupta, Kevin Shah

We are interested in image manipulation via natural language text -- a task that is useful for multiple AI applications but requires complex reasoning over multi-modal spaces. We extend recently proposed Neuro Symbolic Concept Learning (NSCL), which has been quite effective for the task of Visual Question Answering (VQA), for the task of image manipulation.

Matthew Smart, Stanislav Shvartsman, Hayden Nunley

When a founder cell and its progeny divide with incomplete cytokinesis, a network forms in which each intercellular bridge corresponds to a past mitotic event. Networks built in this manner are required for gamete production in many animals, and different species have evolved very different final network topologies. While mechanisms regulating network assemb

Nelson F. Liu, Kenton Lee, Kristina Toutanova

Internet links enable users to deepen their understanding of a topic by providing convenient access to related information. However, the majority of links are unanchored -- they link to a target webpage as a whole, and readers may expend considerable effort localizing the specific parts of the target webpage that enrich their understanding of the link's sour

Fan Bai, Junmo Kang, Gabriel Stanovsky, Dayne Freitag

In this paper, we explore the question of whether large language models can support cost-efficient information extraction from tables. We introduce schema-driven information extraction, a new task that transforms tabular data into structured records following a human-authored schema. To assess various LLM's capabilities on this task, we present a benchmark c

Shuting He, Xudong Jiang, Wei Jiang, Henghui Ding

In this work, we address the challenging task of few-shot and zero-shot 3D point cloud semantic segmentation. The success of few-shot semantic segmentation in 2D computer vision is mainly driven by the pre-training on large-scale datasets like imagenet. The feature extractor pre-trained on large-scale 2D datasets greatly helps the 2D few-shot learning. Howev

Grace Luo, Lisa Dunlap, Dong Huk Park, Aleksander Holynski

Diffusion models have been shown to be capable of generating high-quality images, suggesting that they could contain meaningful internal representations. Unfortunately, the feature maps that encode a diffusion model's internal information are spread not only over layers of the network, but also over diffusion timesteps, making it challenging to extract usefu

James Xu Zhao, Yuxi Xie, Kenji Kawaguchi, Junxian He

Chain-of-Thought (CoT) and Program-Aided Language Models (PAL) represent two distinct reasoning methods, each with its own strengths. CoT employs natural language, offering flexibility and interpretability, while PAL utilizes programming language, yielding more structured and rigorous logic. We introduce a model selection method to combine the best of both w

Benjamin Muller, John Wieting, Jonathan H. Clark, Tom Kwiatkowski

Trustworthy answer content is abundant in many high-resource languages and is instantly accessible through question answering systems, yet this content can be hard to access for those that do not speak these languages. The leap forward in cross-lingual modeling quality offered by generative language models offers much promise, yet their raw generations often

Navita Goyal, Eleftheria Briakou, Amanda Liu, Connor Baumler

NLP systems have shown impressive performance at answering questions by retrieving relevant context. However, with the increasingly large models, it is impossible and often undesirable to constrain models' knowledge or reasoning to only the retrieved context. This leads to a mismatch between the information that the models access to derive the answer and the

Susung Hong, Junyoung Seo, Heeseong Shin, Sunghwan Hong

In the paradigm of AI-generated content (AIGC), there has been increasing attention to transferring knowledge from pre-trained text-to-image (T2I) models to text-to-video (T2V) generation. Despite their effectiveness, these frameworks face challenges in maintaining consistent narratives and handling shifts in scene composition or object placement from a sing

Fivos Kalogiannis, Ioannis Panageas

The works of (Daskalakis et al., 2009, 2022; Jin et al., 2022; Deng et al., 2023) indicate that computing Nash equilibria in multi-player Markov games is a computationally hard task. This fact raises the question of whether or not computational intractability can be circumvented if one focuses on specific classes of Markov games. One such example is two-play

Binwei Yao, Ming Jiang, Tara Bobinac, Diyi Yang

Translating culture-related content is vital for effective cross-cultural communication. However, many culture-specific items (CSIs) often lack viable translations across languages, making it challenging to collect high-quality, diverse parallel corpora with CSI annotations. This difficulty hinders the analysis of cultural awareness of machine translation (M

Da Yin, Xiao Liu, Fan Yin, Ming Zhong

Instruction tuning has emerged to enhance the capabilities of large language models (LLMs) to comprehend instructions and generate appropriate responses. Existing methods either manually annotate or employ LLM (e.g., GPT-series) to generate data for instruction tuning. However, they often overlook associating instructions with existing annotated datasets. In

Lucille Njoo, Chan Young Park, Octavia Stappart, Marvin Thielk

Empowering language is important in many real-world contexts, from education to workplace dynamics to healthcare. Though language technologies are growing more prevalent in these contexts, empowerment has seldom been studied in NLP, and moreover, it is inherently challenging to operationalize because of its implicit nature. This work builds from linguistic a

Yilun Du, Shuang Li, Antonio Torralba, Joshua B. Tenenbaum

Large language models (LLMs) have demonstrated remarkable capabilities in language generation, understanding, and few-shot learning in recent years. An extensive body of work has explored how their performance may be further improved through the tools of prompting, ranging from verification, self-consistency, or intermediate scratchpads. In this paper, we pr

Daniel Deutsch, George Foster, Markus Freitag

Kendall's tau is frequently used to meta-evaluate how well machine translation (MT) evaluation metrics score individual translations. Its focus on pairwise score comparisons is intuitive but raises the question of how ties should be handled, a gray area that has motivated different variants in the literature. We demonstrate that, in settings like modern MT m

Zhipeng Chen, Kun Zhou, Beichen Zhang, Zheng Gong

Although large language models (LLMs) have achieved excellent performance in a variety of evaluation benchmarks, they still struggle in complex reasoning tasks which require specific knowledge and multi-hop reasoning. To improve the reasoning abilities, we propose ChatCoT, a tool-augmented chain-of-thought reasoning framework for chat-based LLMs (e.g., ChatG

Ali Modarressi, Ayyoob Imani, Mohsen Fayyaz, Hinrich Schütze

Large language models (LLMs) have significantly advanced the field of natural language processing (NLP) through their extensive parameters and comprehensive data utilization. However, existing LLMs lack a dedicated memory unit, limiting their ability to explicitly store and retrieve knowledge for various tasks. In this paper, we propose RET-LLM a novel frame

William Brannon, Wonjune Kang, Suyash Fulay, Hang Jiang

Learning on text-attributed graphs (TAGs), in which nodes are associated with one or more texts, has been the subject of much recent work. However, most approaches tend to make strong assumptions about the downstream task of interest, are reliant on hand-labeled data, or fail to equally balance the importance of both text and graph representations. In this w

Vivek Nair, Viktor Radulov, James F. O'Brien

The emergence of affordable standalone virtual reality (VR) devices has allowed VR technology to reach mass-market adoption in recent years, driven primarily by the popularity of VR gaming applications such as Beat Saber. However, despite being the top-grossing VR application to date and the most popular VR e-sport, the population of over 6 million Beat Sabe

Thomas Trogdon

Motivated by Fredholm theory, we develop a framework to establish the convergence of spectral methods for operator equations $\mathcal L u = f$. The framework posits the existence of a left-Fredholm regulator for $\mathcal L$ and the existence of a sufficiently good approximation of this regulator. Importantly, the numerical method itself need not make use o

Cheng Qian, Chi Han, Yi R. Fung, Yujia Qin

Large Language Models (LLMs) have made significant progress in utilizing tools, but their ability is limited by API availability and the instability of implicit reasoning, particularly when both planning and execution are involved. To overcome these limitations, we propose CREATOR, a novel framework that enables LLMs to create their own tools using documenta

Kate Napier, Mike Gladders, Keren Sharon, Håkon Dahle

We report the discovery of COOL J0335$-$1927, a quasar at z = 3.27 lensed into three images with a maximum separation of 23.3" by a galaxy cluster at z = 0.4178. To date this is the highest redshift wide-separation lensed quasar known. In addition, COOL J0335$-$1927 shows several strong intervening absorbers visible in the spectra of all three quasar images

N. Akchurin, G. Altopp, B. Burkle, W. D. Frey

Surface damage caused by ionizing radiation in SiO$_2$ passivated silicon particle detectors consists mainly of the accumulation of a positively charged layer along with trapped-oxide-charge and interface traps inside the oxide and close to the Si/SiO$_2$-interface. High density positive interface net charge can be detrimental to the operation of a multi-cha

Maximilian F. Steffen

Existing results for the estimation of the L\'evy measure are mostly limited to the onedimensional setting. We apply the spectral method to multidimensional L\'evy processes in order to construct a nonparametric estimator for the multivariate jump distribution. We prove convergence rates for the uniform estimation error under both a low- and a high-frequency

Tim Dettmers, Artidoro Pagnoni, Ari Holtzman, Luke Zettlemoyer

We present QLoRA, an efficient finetuning approach that reduces memory usage enough to finetune a 65B parameter model on a single 48GB GPU while preserving full 16-bit finetuning task performance. QLoRA backpropagates gradients through a frozen, 4-bit quantized pretrained language model into Low Rank Adapters~(LoRA). Our best model family, which we name Guan

Katharina Laubscher, Jay D. Sau, Sankar Das Sarma

We theoretically study gate-defined one-dimensional channels in planar Ge hole gases as a potential platform for non-Abelian Majorana zero modes. We model the valence band holes in the Ge channel by adding appropriate confinement potentials to the 3D Luttinger-Kohn Hamiltonian, additionally taking into account a magnetic field applied parallel to the channel

Tsu-Jui Fu, Wenhan Xiong, Yixin Nie, Jingyu Liu

3D human modeling has been widely used for engaging interaction in gaming, film, and animation. The customization of these characters is crucial for creativity and scalability, which highlights the importance of controllability. In this work, we introduce Text-guided 3D Human Generation (\texttt{T3H}), where a model is to generate a 3D human, guided by the f

Alkis Kalavasis, Amin Karbasi, Shay Moran, Grigoris Velegkas

When two different parties use the same learning rule on their own data, how can we test whether the distributions of the two outcomes are similar? In this paper, we study the similarity of outcomes of learning rules through the lens of the Total Variation (TV) distance of distributions. We say that a learning rule is TV indistinguishable if the expected TV

Yida Mu, Ben P. Wu, William Thorne, Ambrose Robinson

Instruction-tuned Large Language Models (LLMs) have exhibited impressive language understanding and the capacity to generate responses that follow specific prompts. However, due to the computational demands associated with training these models, their applications often adopt a zero-shot setting. In this paper, we evaluate the zero-shot performance of two pu