May 2023 arXiv papers — page 78

Zhisong Zhang, Emma Strubell, Eduard Hovy

In this work we propose a pragmatic method that reduces the annotation cost for structured label spaces using active learning. Our approach leverages partial annotation, which reduces labeling costs for structured outputs by selecting only the most informative sub-structures for annotation. We also utilize self-training to incorporate the current model's aut

Jiayu Chen, Dipesh Tamboli, Tian Lan, Vaneet Aggarwal

Multi-task Imitation Learning (MIL) aims to train a policy capable of performing a distribution of tasks based on multi-task expert demonstrations, which is essential for general-purpose robots. Existing MIL algorithms suffer from low data efficiency and poor performance on complex long-horizontal tasks. We develop Multi-task Hierarchical Adversarial Inverse

Masato Hisakado, Takuya Kaneko

We study the eigenvalue of the Wishart matrix, which is created from a time series with temporal correlation. When there is no correlation, the eigenvalue distribution of the Wishart matrix is known as the Marchenko-Pastur distribution (MPD) in the double scaling limit. When there is temporal correlation, the eigenvalue distribution converges to the deformed

Feng Wang, Chuan-Fu Yang

In this work, we consider Dirac-type operators with a constant delay less than two-fifths of the interval and not less than one-third of the interval. For our considered Dirac-type operators, an incomplete inverse spectral problem is studied. Specifically, when two complex potentials are known a priori on a certain subinterval, reconstruction of the two pote

Xiangjun Wang, Yu Zhang

This paper studies the higher differentials of the classical Adams spectral sequence at odd primes. In particular, we follow the ``cofiber of $\tau$ philosophy'' of Gheorghe, Isaksen, Wang, and Xu to show that higher Adams differentials agree with their corresponding higher algebraic Novikov differentials in a certain range.

Sanghoon Lee

In this paper, we establish an improved decay estimate for the Dirichlet energy of Dir-stationary $Q$-valued functions. As a direct application of this estimate, we derive a Liouville-type theorem for bounded Dir-stationary $Q$-valued functions defined on $\mathbb{R}^m$. Additionally, in an attempt to establish the continuity of Dir-stationary $Q$-valued fun

Xianchao Wu

Speech-to-speech translation is a typical sequence-to-sequence learning task that naturally has two directions. How to effectively leverage bidirectional supervision signals to produce high-fidelity audio for both directions? Existing approaches either train two separate models or a multitask-learned model with low efficiency and inferior performance. In thi

Zhibin Gou, Qingyan Guo, Yujiu Yang

Generative methods greatly promote aspect-based sentiment analysis via generating a sequence of sentiment elements in a specified format. However, existing studies usually predict sentiment elements in a fixed order, which ignores the effect of the interdependence of the elements in a sentiment tuple and the diversity of language expression on the results. I

Walter Goodwin, Ioannis Havoutis, Ingmar Posner

In order to meaningfully interact with the world, robot manipulators must be able to interpret objects they encounter. A critical aspect of this interpretation is pose estimation: inferring quantities that describe the position and orientation of an object in 3D space. Most existing approaches to pose estimation make limiting assumptions, often working only

Erina Yamaguchi, Sai Ravela

Nonlinear receding horizon model predictive control is a powerful approach to controlling nonlinear dynamical systems. However, typical approaches that use the Jacobian, adjoint, and forward-backward passes may lose fidelity and efficacy for highly nonlinear problems. Here, we develop an Ensemble Model Predictive Control (EMPC) approach wherein the forward m

Yuanyuan Luan, Roger S. Zoh, Erjia Cui, Xue Lan

Wearable devices permit the continuous monitoring of biological processes, such as blood glucose metabolism, and behavior, such as sleep quality and physical activity. The continuous monitoring often occurs in epochs of 60 seconds over multiple days, resulting in high dimensional longitudinal curves that are best described and analyzed as functional data. Fr

Archana Vadakattu, Michelle Blom, Adrian R. Pearce

The ability to continuously learn and adapt to new situations is one where humans are far superior compared to AI agents. We propose an approach to knowledge transfer using behavioural strategies as a form of transferable knowledge influenced by the human cognitive ability to develop strategies. A strategy is defined as a partial sequence of events - where a

Ming Ying Yang, Gloria Hyunjung Kwak, Tom Pollard, Leo Anthony Celi

Social determinants of health (SDOH) -- the conditions in which people live, grow, and age -- play a crucial role in a person's health and well-being. There is a large, compelling body of evidence in population health studies showing that a wide range of SDOH is strongly correlated with health outcomes. Yet, a majority of the risk prediction models based on

Ali Ayub, Zachary De Francesco, Patrick Holthaus, Chrystopher L. Nehaniv

For long-term deployment in dynamic real-world environments, assistive robots must continue to learn and adapt to their environments. Researchers have developed various computational models for continual learning (CL) that can allow robots to continually learn from limited training data, and avoid forgetting previous knowledge. While these CL models can miti

Ashish Sinha, Jeremy Kawahara, Arezou Pakzad, Kumar Abhishek

In recent years, deep learning (DL) has shown great potential in the field of dermatological image analysis. However, existing datasets in this domain have significant limitations, including a small number of image samples, limited disease conditions, insufficient annotations, and non-standardized image acquisitions. To address these shortcomings, we propose

Ioana Baldini, Chhavi Yadav, Manish Nagireddy, Payel Das

Bias auditing of language models (LMs) has received considerable attention as LMs are becoming widespread. As such, several benchmarks for bias auditing have been proposed. At the same time, the rapid evolution of LMs can make these benchmarks irrelevant in no time. Bias auditing is further complicated by LM brittleness: when a presumably biased outcome is o

Yaping Sun, Hao Chen, Xiaodong Xu, Ping Zhang

Remote zero-shot object recognition, i.e., offloading zero-shot object recognition task from one mobile device to remote mobile edge computing (MEC) server or another mobile device, has become a common and important task to solve for 6G. In order to tackle this problem, this paper first establishes a zero-shot multi-level feature extractor, which projects th

Jiahao Chen, Yurou Liu, Jiangmeng Li, Bing Su

Molecular representation learning is a crucial task in predicting molecular properties. Molecules are often modeled as graphs where atoms and chemical bonds are represented as nodes and edges, respectively, and Graph Neural Networks (GNNs) have been commonly utilized to predict atom-related properties, such as reactivity and solubility. However, functional g

Zihao Chen, Jia Lu, Xing-Ming Zhao, Haiyang Yu

Cancer is a systemic heterogeneous disease involving complex molecular networks. Tumor formation involves epithelial-mesenchymal transition (EMT), which promotes both metastasis and plasticity of cancer cells. Recent experiments proposed that cancer cells can be transformed into adipocytes with combination drugs. However, the underlying mechanisms for how th

Isaac Gibbs, John J. Cherian, Emmanuel J. Candès

We consider the problem of constructing distribution-free prediction sets with finite-sample conditional guarantees. Prior work has shown that it is impossible to provide exact conditional coverage universally in finite samples. Thus, most popular methods only guarantee marginal coverage over the covariates or are restricted to a limited set of conditional t

Gui-Qiang G. Chen, Feimin Huang, Tianhong Li, Weiqiang Wang

We are concerned with global finite-energy solutions of the three-dimensional compressible Euler-Poisson equations with gravitational potential and general pressure law, especially including the constitutive equation of white dwarf stars. We construct global finite-energy solutions of the Cauchy problem for the Euler-Poisson equations with large initial data

Yaohui Guo, X. Jessie Yang, Cong Shi

Trust has been identified as a central factor for effective human-robot teaming. Existing literature on trust modeling predominantly focuses on dyadic human-autonomy teams where one human agent interacts with one robot. There is little, if not no, research on trust modeling in teams consisting of multiple human agents and multiple robotic agents. To fill thi

Oliver Knill

If G is a finite abstract simplicial complex and K is a subcomplex of G and U=G-K is the open complement of K in G, the Betti vectors of K and U and G satisfy the inequality b(G) less or equal b(K)+b(U).

Luke Gessler

Evaluation datasets are critical resources for measuring the quality of pretrained language models. However, due to the high cost of dataset annotation, these resources are scarce for most languages other than English, making it difficult to assess the quality of language models. In this work, we present a new method for evaluation dataset construction which

Jun-Jin Peng, Yao Wang, Wei-Jie Guo

We investigate the conserved quantities associated to Killing isometries for asymptotically AdS spacetimes within the framework of quadratic-curvature gravity. By constructing a rank-4 tensor possessing the same index symmetries as the ones of the Riemann tensor, we propose a 2-form potential resembling the Noether one for quadratic-curvature gravity. Such a

J. S. Porto, A. R. Vieira

We compute the classical and the quantum breaking of the dilatation current in the minimal Lorentz and CPT-violating quantum electrodynamics. At the classical level, scale symmetry is broken by the general mass term \bar{\psi}M\psi and the Chern-Simons-like term. At the quantum level, it is broken by all expected observable fermion operators in the massless

Sang Hyun Park, Michael Sammon, Eugene Mele, Tony Low

Artificial lattices have been used as a platform to extend the application of topological physics beyond electronic systems. Here, using the two-dimensional Lieb lattice as a prototypical example, we show that an array of disks which each support localized plasmon modes give rise to an analog of the quantum spin Hall state enforced by a synthetic time revers

Raf Bocklandt, Jasper van de Kreeke

Mirror symmetry originally envisions a correspondence between deformations of the A-side and deformations of the B-side. In this paper, we achieve an explicit correspondence in the case of punctured surfaces. The starting point is the noncommutative mirror equivalence $ \operatorname{Gtl} Q \cong \operatorname{mf} (\operatorname{Jac} \check{Q}, \ell) $ for a

Panayiotis Danassis, Shresth Verma, Jackson A. Killian, Aparna Taneja

The success of many healthcare programs depends on participants' adherence. We consider the problem of scheduling interventions in low resource settings (e.g., placing timely support calls from health workers) to increase adherence and/or engagement. Past works have successfully developed several classes of Restless Multi-armed Bandit (RMAB) based soluti

Mazhar Chebl, Xing He, Ding-Shyue Yang

Revived attention in black phosphorus (bP) has been tremendous in the past decade. While many photoinitiated experiments have been conducted, a cross-examination of bP's photocarrier and structural dynamics is still lacking. In this report, we provide such analysis by examining time-resolved data acquired using optical transient reflectivity and reflecti

Antoine Oustry, Matteo Tacchi, Didier Henrion

The Lasserre or moment-sum-of-square hierarchy of linear matrix inequality relaxations is used to compute inner approximations of the maximal positively invariant set for continuous-time dynamical systems with polynomial vector fields. Convergence in volume of the hierarchy is proved under a technical growth condition on the average exit time of trajectories

Ian Hambleton, Jonathan A. Hillman

We consider closed topological 4-manifolds $M$ with universal cover ${S^2\times{S^2}}$ and Euler characteristic $χ(M) = 1$. All such manifolds with $π=π_1(M)\cong {\mathbb Z}/4$ are homotopy equivalent. In this case, we show that there are four homeomorphism types, and propose a candidate for a smooth example which is not homeomorphic to the geometric quotie

Drew A. Geller, Johanna L. Mathieu

Driven by the need to offset the variability of wind and solar generation on the electrical grid, development of load controls is a highly active field in the engineering literature. However, practical use of residential loads for grid balancing and ancillary services remains rare, in part due to the relative cost of communicating with hordes of small loads

Andrew Rouditchenko, Sameer Khurana, Samuel Thomas, Rogerio Feris

Recent models such as XLS-R and Whisper have made multilingual speech technologies more accessible by pre-training on audio from around 100 spoken languages each. However, there are thousands of spoken languages worldwide, and adapting to new languages is an important problem. In this work, we aim to understand which model adapts better to languages unseen d

Daniel Fernandes Gomes, Paolo Paoletti, Shan Luo

Recently, several morphologies, each with its advantages, have been proposed for the \textit{GelSight} high-resolution tactile sensors. However, existing simulation methods are limited to flat-surface sensors, which prevents their usage with the newer sensors of non-flat morphologies in Sim2Real experiments. In this paper, we extend a previously proposed Gel

Takashi Ishizuka

This paper proposes a new one-sided matching market model in which every agent has a cost function that is allowed to take a negative value. Our model aims to capture the situation where some agents can profit by exchanging their obtained goods with other agents. We formulate such a model based on a graphical one-sided matching market, introduced by Massand

Meng Li, Paul Grigas, Alper Atamturk

We study a new penalty reformulation of constrained convex optimization based on the softplus penalty function. We develop novel and tight upper bounds on the objective value gap and the violation of constraints for the solutions to the penalty reformulations by analyzing the solution path of the reformulation with respect to the smoothness parameter. We use

Justin Eilertsen, Santiago Schnell, Sebastian Walcher

There is a vast amount of literature concerning the appropriateness of various perturbation parameters for the standard quasi-steady state approximation in the Michaelis-Menten reaction mechanism, and also concerning the relevance of these parameters for the accuracy of the approximation by the familiar Michaelis-Menten equation. Typically, the arguments in

Lê Thành Dũng Nguyên

We consider the following decision problem: given two simply typed $\lambda$-terms, are they $\beta$-convertible? Equivalently, do they have the same normal form? It is famously non-elementary, but the precise complexity - namely TOWER-complete - is lesser known. One goal of this short paper is to popularize this fact. Our original contribution is to show th

Qian Huang, Hongyu Ren, Peng Chen, Gregor Kržmanc

In-context learning is the ability of a pretrained model to adapt to novel and diverse downstream tasks by conditioning on prompt examples, without optimizing any parameters. While large language models have demonstrated this ability, how in-context learning could be performed over graphs is unexplored. In this paper, we develop \textbf{Pr}etraining \textbf{

Qiming Bao, Alex Yuxuan Peng, Zhenyun Deng, Wanjun Zhong

Combining large language models with logical reasoning enhances their capacity to address problems in a robust and reliable manner. Nevertheless, the intricate nature of logical reasoning poses challenges when gathering reliable data from the web to build comprehensive training datasets, subsequently affecting performance on downstream tasks. To address this

Takahiro Ueda, Satoshi Okuzumi, Akimasa Kataoka, Mario Flock

Midplane heating induced by disk accretion plays a key role in determining the disk temperature particularly at the inner disk midplane where planets form. However, the efficiency of accretion heating has been not well constrained by observations. We construct two-dimensional models of the Class II disk around CW Tau, taking into account the midplane heating

Jinglei Cheng, Zhiding Liang, Rui Yang, Hang Ren

Most previous research focused on designing pulse programs without considering the performance of individual elements or the final fidelity. To evaluate the performance of quantum pulses, it is required to know the noiseless results of the pulses. However, quantum pulses can implement unitary matrices that are not analytically known to the user, and pulse si

Shivangi Yadav, Arun Ross

Generative Adversarial Networks (GANs) have shown success in approximating complex distributions for synthetic image generation. However, current GAN-based methods for generating biometric images, such as iris, have certain limitations: (a) the synthetic images often closely resemble images in the training dataset; (b) the generated images lack diversity in

Muhammad Abdullah Hanif, Muhammad Shafique

Fault-aware retraining has emerged as a prominent technique for mitigating permanent faults in Deep Neural Network (DNN) hardware accelerators. However, retraining leads to huge overheads, specifically when used for fine-tuning large DNNs designed for solving complex problems. Moreover, as each fabricated chip can have a distinct fault pattern, fault-aware r

Fanghua Ye, Zhiyuan Hu, Emine Yilmaz

Dialogue systems have received increasing attention while automatically evaluating their performance remains challenging. User satisfaction estimation (USE) has been proposed as an alternative. It assumes that the performance of a dialogue system can be measured by user satisfaction and uses an estimator to simulate users. The effectiveness of USE depends he

Djordje Minic

This talk summarizes a new understanding of the cosmological constant problem, which essentially relies on a phase-space-like computation of the vacuum energy, both in the realm of quantum field theory coupled to gravity, and in the realm of a consistent formulation of a quantum theory of gravity and matter, such as string theory, combined with central prope

Eric Tan, R. Ganesh

Scattering off a potential is a fundamental problem in quantum physics. It has been studied extensively with amplitudes derived for various potentials. In this article, we explore a setting with no potentials, where scattering occurs off a junction where many wires meet. We study this problem using a tight-binding discretization of a star graph geometry -- o

Noah P. Baker, Valeri P. Frolov

In this paper, the orbits of a charged particle near the event horizon of a magnetized black hole are investigated. For a static black hole of mass $M$ immersed in a homogeneous magnetic field $B$, the dimensionless parameter $b=eBGM/ (mc^4)$ controls the radius of the circular orbits and determines the position of the innermost stable circular orbit (ISCO),

Muhammad Abdullah Hanif, Muhammad Shafique

Permanent faults induced due to imperfections in the manufacturing process of Deep Neural Network (DNN) accelerators are a major concern, as they negatively impact the manufacturing yield of the chip fabrication process. Fault-aware training is the state-of-the-art approach for mitigating such faults. However, it incurs huge retraining overheads, specificall

Antoine Detaille

We consider the problem of strong density of smooth maps in the Sobolev space $ W^{s,p}(Q^{m};\mathcal{N}) $, where $ 0 < s < +\infty $, $ 1 \leq p < +\infty $, $ Q^{m} $ is the unit cube in $ \mathbb{R}^{m} $, and $ \mathcal{N} $ is a smooth compact connected Riemannian manifold without boundary. Our main result fully answers the strong density problem in t

Juanjuan Lu, Danila N. Puzyrev, Vladislav V. Pankratov, Dmitry V. Skryabin

Frequency conversion of dissipative solitons associated with the generation of broadband optical frequency combs having a tooth spacing of hundreds of giga-hertz is a topical challenge holding the key to practical applications in precision spectroscopy and data processing. The work in this direction is underpinned by fundamental problems in nonlinear and qua

Katerina Margatina, Nikolaos Aletras

Active learning (AL) is a human-and-model-in-the-loop paradigm that iteratively selects informative unlabeled data for human annotation, aiming to improve over random sampling. However, performing AL experiments with human annotations on-the-fly is a laborious and expensive process, thus unrealistic for academic research. An easy fix to this impediment is to

Ravi Shankar, Yu Yuan

We derive a priori interior Hessian estimates and interior regularity for the $\sigma_2$ equation in dimension four. Our method provides respectively a new proof for the corresponding three dimensional results and a Hessian estimate for smooth solutions satisfying a dynamic semi-convexity condition in higher $n\ge 5$ dimensions.

Linyong Nan, Yilun Zhao, Weijin Zou, Narutatsu Ri

In-context learning (ICL) has emerged as a new approach to various natural language processing tasks, utilizing large language models (LLMs) to make predictions based on context that has been supplemented with a few examples or task-specific instructions. In this paper, we aim to extend this method to question answering tasks that utilize structured knowledg

Wilson G. Gregory, David W. Hogg, Ben Blum-Smith, Maria Teresa Arias

Machine learning methods are increasingly being employed as surrogate models in place of computationally expensive and slow numerical integrators for a bevy of applications in the natural sciences. However, while the laws of physics are relationships between scalars, vectors, and tensors that hold regardless of the frame of reference or chosen coordinate sys

Rui Wang, Yuesheng Xu, Mingsong Yan

Sparsity of a learning solution is a desirable feature in machine learning. Certain reproducing kernel Banach spaces (RKBSs) are appropriate hypothesis spaces for sparse learning methods. The goal of this paper is to understand what kind of RKBSs can promote sparsity for learning solutions. We consider two typical learning models in an RKBS: the minimum norm

Ahsan Mehmood, Asma Sarauji, M. Mahboob Ur Rahman, Tareq Y. Al-Naffouri

In the post-covid19 era, every new wave of the pandemic causes an increased concern among the masses to learn more about their state of well-being. Therefore, it is the need of the hour to come up with ubiquitous, low-cost, non-invasive tools for rapid and continuous monitoring of body vitals that reflect the status of one's overall health. In this backdrop,

Stephen J. Dilworth, Denka Kutzarova, Mikhail I. Ostrovskii

The paper starts with discussion of applications of cycle spaces to transportation cost. After a short survey of the known results on cycle spaces, we turn to the study of minimal projections onto cycle spaces in the corresponding $\ell_1$-spaces. This study is naturally related to the study of invariant projections on the cycle space, which, in turn, are de

Erik Drysdale

Post-selection inference (PoSI) is a statistical technique for obtaining valid confidence intervals and p-values when hypothesis generation and testing use the same source of data. PoSI can be used on a range of popular algorithms including the Lasso. Data carving is a variant of PoSI in which a portion of held out data is combined with the hypothesis genera

Marc E. Canby, Julia Hockenmaier

Transformer-based encoder-decoder models that generate outputs in a left-to-right fashion have become standard for sequence-to-sequence tasks. In this paper, we propose a framework for decoding that produces sequences from the "outside-in": at each step, the model chooses to generate a token on the left, on the right, or join the left and right sequences. We

Karel Beneš, Martin Kocour, Lukáš Burget

End-to-end (e2e) systems have recently gained wide popularity in automatic speech recognition. However, these systems do generally not provide well-calibrated word-level confidences. In this paper, we propose Hystoc, a simple method for obtaining word-level confidences from hypothesis-level scores. Hystoc is an iterative alignment procedure which turns hypot

Huaisheng Zhu, Dongsheng Luo, Xianfeng Tang, Junjie Xu

Graph Neural Networks (GNNs) have achieved state-of-the-art performance for link prediction. However, GNNs suffer from poor interpretability, which limits their adoptions in critical scenarios that require knowing why certain links are predicted. Despite various methods proposed for the explainability of GNNs, most of them are post-hoc explainers developed f

Korrawe Karunratanakul, Konpat Preechakul, Supasorn Suwajanakorn, Siyu Tang

Denoising diffusion models have shown great promise in human motion synthesis conditioned on natural language descriptions. However, integrating spatial constraints, such as pre-defined motion trajectories and obstacles, remains a challenge despite being essential for bridging the gap between isolated human motion and its surrounding environment. To address

Rami Aly, Xingjian Shi, Kaixiang Lin, Aston Zhang

A particularly successful class of approaches for few-shot learning combines language models with prompts -- hand-crafted task descriptions that complement data samples. However, designing prompts by hand for each task commonly requires domain knowledge and substantial guesswork. We observe, in the context of classification tasks, that instruction finetuned

Scott K. Hansen, Yichen Tao, Satish Karra

Motivated by CO2 capture and sequestration (CCS) design considerations, we consider the coupled effects of permeability heterogeneity and background flow on the dissolution of a supercritical CO2 lens into an underlying deep, confined aquifer. We present the results of a large-scale Monte Carlo simulation study examining the interaction of background flow ra

Subham Sahoo, Arpan Malkhandi, Kristian Skafte Jensen

In this article, we determine a fundamental anatomical modeling parallelism between low-inertia power systems and Bohr's atomic model. The proposed atomic architecture will serve as a microscopic building block, where we validate the structural analogy of low-inertia power systems using semi-classical quantum approximations in IEEE 9-bus & 39-bus systems. As

J. W. Zhou, F. Wyrowski, S. Neupane, J. S. Urquhart

Hub-filament systems are suggested to be the birth cradles of high-mass stars and clusters. We apply the FILFINDER algorithm to the integrated intensity maps of the 13CO (3-2) line to identify filaments in the G333 complex, and extract the velocity and intensity along the filament skeleton from moment maps. Clear velocity and density fluctuations are seen al

Xingting Gong, Manu Prakash

Activity can organize matter in unique configurations inaccessible to equilibrium systems, including a sundry of spiraling shapes seen in nature that range from galaxies to living tissues to fossilized stromatolites. How these dynamic yet stable patterns form in motile active systems that span a range of length and time scales remains an open question. Here

Iordanis Fostiropoulos, Bowman Brown, Laurent Itti

Machine learning is facing a 'reproducibility crisis' where a significant number of works report failures when attempting to reproduce previously published results. We evaluate the sources of reproducibility failures using a meta-analysis of 142 replication studies from ReScience C and 204 code repositories. We find that missing experiment details such as hy

Luuk Jacobs, Stefano Mandija, Hongyan Liu, Cornelis A. T. van den Berg

In this study, we develop a physics-informed deep learning-based method to synthesize multiple brain magnetic resonance imaging (MRI) contrasts from a single five-minute acquisition and investigate its ability to generalize to arbitrary contrasts to accelerate neuroimaging protocols. A dataset of fifty-five subjects acquired with a standard MRI protocol and

Zheng Dong, Zekai Fan, Shixiang Zhu

Point processes offer a versatile framework for sequential event modeling. However, the computational challenges and constrained representational power of the existing point process models have impeded their potential for wider applications. This limitation becomes especially pronounced when dealing with event data that is associated with multi-dimensional o

Hanmin Li, Avetik Karagulyan, Peter Richtárik

This paper introduces a new method for minimizing matrix-smooth non-convex objectives through the use of novel Compressed Gradient Descent (CGD) algorithms enhanced with a matrix-valued stepsize. The proposed algorithms are theoretically analyzed first in the single-node and subsequently in the distributed settings. Our theoretical results reveal that the ma

Linyuan Gong, Chenyan Xiong, Xiaodong Liu, Payal Bajaj

This paper explores the effectiveness of model-generated signals in improving zero-shot generalization of text-to-text Transformers such as T5. We study various designs to pretrain T5 using an auxiliary model to construct more challenging token replacements for the main model to denoise. Key aspects under study include the decoding target, the location of th

Bartosz Fornal, Kassandra Garcia, Erika Pierre

We propose to search for a new type of gravitational wave signature relevant for particle physics models with symmetries broken at vastly different energy scales. The spectrum contains a characteristic double-peak structure consisting of a sharp peak from domain walls and a smooth bump from a first order phase transition in the early Universe. We demonstrate

Ziqi Wang, Chi Han, Wenxuan Bao, Heng Ji

Knowledge distillation (KD) requires sufficient data to transfer knowledge from large-scale teacher models to small-scale student models. Therefore, data augmentation has been widely used to mitigate the shortage of data under specific scenarios. Classic data augmentation techniques, such as synonym replacement and k-nearest-neighbors, are initially designed

Jared Wong, Jin Kim

We investigate how people perceive ChatGPT, and, in particular, how they assign human-like attributes such as gender to the chatbot. Across five pre-registered studies (N = 1,552), we find that people are more likely to perceive ChatGPT to be male than female. Specifically, people perceive male gender identity (1) following demonstrations of ChatGPT's core a

Jordan Meadows, Marco Valentino, Damien Teney, Andre Freitas

This paper proposes a methodology for generating and perturbing detailed derivations of equations at scale, aided by a symbolic engine, to evaluate the generalisability of Transformers to out-of-distribution mathematical reasoning problems. Instantiating the framework in the context of sequence classification tasks, we compare the capabilities of GPT-4, GPT-

Claire Alamichel, Juan Calvo, Erwan Hingant, Saoussen Latrach

We present a new modeling approach for G protein coupled receptors signaling systems, that take into account the compartmentalization of receptors and their effectors, both at plasma membrane and in dynamic intra-cellular vesicles called endosomes. The first building block of the model is about compartment dynamics. It takes into account creation of de-novo

Álvaro Becerra, Roberto Daza, Ruth Cobos, Aythami Morales

In this article, we present a Web-based System called M2LADS, which supports the integration and visualization of multimodal data recorded in learning sessions in a MOOC in the form of Web-based Dashboards. Based on the edBB platform, the multimodal data gathered contains biometric and behavioral signals including electroencephalogram data to measure learner

Juan Calvo, Erwan Hingant, Romain Yvinec

We consider the Lifshitz-Slyozov model with inflow boundary conditions of nucleation type. We show that for a collection of representative rate functions the size distributions approach degenerate states concentrated at zero size for sufficiently large times. The proof relies on monotonicity properties of some quantities associated to an entropy functional.

Zsolt Pocze

In this paper, we present a new multi-scale information content calculation method based on Shannon information (and Shannon entropy). The original method described by Claude E. Shannon and based on the logarithm of the probability of elements gives an upper limit to the information content of discrete patterns, but in many cases (for example, in the case of

Ada Stelzer, Alexander Yong

Abhyankar defined an ideal to be Hilbertian if its Hilbert polynomial coincides with its Hilbert function for all nonnegative integers. In 1984, he proved that the ideal of (r+1)-order minors of a generic p x q matrix is Hilbertian. We give a different proof and a generalization to the Schubert determinantal ideals introduced by Fulton in 1992. Our proof red

Junyi Zhu, Xingchen Ma, Matthew B. Blaschko

Federated Learning (FL) is a distributed learning scheme to train a shared model across clients. One common and fundamental challenge in FL is that the sets of data across clients could be non-identically distributed and have different sizes. Personalized Federated Learning (PFL) attempts to solve this challenge via locally adapted models. In this work, we p

S. Michael Fall, Vicente Rodriguez-Gomez

In cosmological simulations without baryons, the relation between the specific angular momentum $j_{\rm h}$ and mass $M_{\rm h}$ of galactic dark matter halos has the well-established form $j_{\rm h} \propto M_{\rm h}^{2/3}$. This is invariably adopted as the starting point in efforts to understand the analogous relation between the specific angular momentum

Pablo Portilla Cuadrado, Baldur Sigurðsson

For any plane curve singularity defined by an analytic function germ $f$, we construct a spine on each Milnor fiber simultaneously, that realizes the vanishing topology. In order to do so, we study the separatrices at the origin of the vector field $-\nabla \log |f|$. Under some genericity conditions on the metric, we produce a natural partition of the set o

Jiarui Sun, Girish Chowdhary

Stochastic Human Motion Prediction (HMP) aims to predict multiple possible future human pose sequences from observed ones. Most prior works learn motion distributions through encoding-decoding in the latent space, which does not preserve motion's spatial-temporal structure. While effective, these methods often require complex, multi-stage training and yield

Xin Guo, Xinyu Li, Chinmay Maheshwari, Shankar Sastry

We propose a new framework of Markov $\alpha$-potential games to study Markov games. We show that any Markov game with finite-state and finite-action is a Markov $\alpha$-potential game, and establish the existence of an associated $\alpha$-potential function. Any optimizer of an $\alpha$-potential function is shown to be an $\alpha$-stationary Nash equilibr

Huadai Liu, Rongjie Huang, Jinzheng He, Gang Sun

Speech-to-SQL (S2SQL) aims to convert spoken questions into SQL queries given relational databases, which has been traditionally implemented in a cascaded manner while facing the following challenges: 1) model training is faced with the major issue of data scarcity, where limited parallel data is available; and 2) the systems should be robust enough to handl

Gustau Camps-Valls, Andreas Gerhardus, Urmi Ninad, Gherardo Varando

Physics is a field of science that has traditionally used the scientific method to answer questions about why natural phenomena occur and to make testable models that explain the phenomena. Discovering equations, laws and principles that are invariant, robust and causal explanations of the world has been fundamental in physical sciences throughout the centur

Xiaoda Qu, Xiran Fan, Baba C. Vemuri

Distributional approximation is a fundamental problem in machine learning with numerous applications across all fields of science and engineering and beyond. The key challenge in most approximation methods is the need to tackle the intractable normalization constant pertaining to the parametrized distributions used to model the data. In this paper, we presen

Gaosheng Liu, Lin Wang

Recently, intermittent computing (IC) has received tremendous attention due to its high potential in perpetual sensing for Internet-of-Things (IoT). By harvesting ambient energy, battery-free devices can perform sensing intermittently without maintenance, thus significantly improving IoT sustainability. To build a practical intermittently-powered sensing sys

Marco Piva

We review the formulation of quantum field theories with purely virtual particles, a new type of degrees of freedom that can mediate interactions without ever appear as external on-shell states. This property allows to solve the problem of ghosts in higher-derivative quantum gravity, leading to a renormalizable and unitary theory. The main steps for the BRST

Daniel Minahan

We show that the complex of homologous curves of a closed, oriented surface of genus g is (g-3)--acyclic.

Calin Iuliu Lazaroiu

We construct natural local coordinate systems on the phase space of two-field cosmological models with orientable target space, which allow for a description of cosmological flows through quantities of direct physical interest. Such coordinates are induced by the fundamental observables of the model, which we formulate geometrically using the tautological bu

Bulent Sagir, Erdogan Aydin, Haci Ilhan

This letter proposes a novel deep neural network (DNN) assisted cooperative reconfigurable intelligent surface (RIS) scheme and a DNN-based symbol detection model for intervehicular communication over cascaded Nakagami-m fading channels. In the considered realistic channel model, the channel links between moving nodes are modeled as cascaded Nakagami-m chann

Jafar Sadeghi, Saeed Noori Gashti, Mohammad Reza Alipour, Mohammad Ali S. Afshar

Recently, it was shown in [1] that black holes are the source of dark energy. In [2-5], the truth or falsity of this concept has been discussed. We briefly state the arguments raised in each of these papers, but our main goal is not to accept or reject these debates. Rather, in this note, we show that black holes in specific structures, such as the Reissner-

Oana Ignat, Zhijing Jin, Artem Abzaliev, Laura Biester

Recent progress in large language models (LLMs) has enabled the deployment of many generative NLP applications. At the same time, it has also led to a misleading public discourse that ``it's all been solved.'' Not surprisingly, this has, in turn, made many NLP researchers -- especially those at the beginning of their careers -- worry about what NLP research

Ibrahim Ahmed, Marcos Quinones-Grueiro, Gautam Biswas

In this work, we present an approach to supervisory reinforcement learning control for unmanned aerial vehicles (UAVs). UAVs are dynamic systems where control decisions in response to disturbances in the environment have to be made in the order of milliseconds. We formulate a supervisory control architecture that interleaves with extant embedded control and

Zachary Yang, Yasmine Maricar, MohammadReza Davari, Nicolas Grenon-Godbout

Detecting toxicity in online spaces is challenging and an ever more pressing problem given the increase in social media and gaming consumption. We introduce ToxBuster, a simple and scalable model trained on a relatively large dataset of 194k lines of game chat from Rainbow Six Siege and For Honor, carefully annotated for different kinds of toxicity. Compared