October 2024 arXiv papers — page 22

Efstratios Manousakis

We argue that alternating-layer structures of lattice mismatched or misaligned (twisted) atomically-thin layers should be expected to be more efficient absorbers of the broad-spectrum of solar radiation than the bulk material of each individual layer. In such mismatched layer-structures the conduction and valence bands of the bulk material, split into multip

Grace Luo, Trevor Darrell, Amir Bar

Autoregressive vision-language models (VLMs) can handle many tasks within a single model, yet the representations that enable this capability remain opaque. We find that VLMs align conceptually equivalent inputs into a shared task vector, which is invariant to modality (text, image) and format (examples, instruction), and may simplify VLM processing. We meas

Elena Hoster

We provide explicit combinatorial formulas for the Chow polynomial and for the augmented Chow polynomial of uniform matroids, thereby proving a conjecture by Ferroni. These formulas refine existing formulas by Hampe and by Eur, Huh, and Larson, offering a combinatorial interpretation of the coefficients based on Schubert matroids. As a byproduct, we count Sc

Kaif Hilman, Sil Linskens

We lay down the foundations of a theory of parametrised functor calculus, generalising parts of the functor calculus of Goodwillie. We introduce the notion of excisable posets and develop a theory of excisive approximations in this context. As an application, we introduce two different excisable posets when parametrising over an atomic orbital category. By c

Erik Rydow, Vijay P. Singh, Abel Beregi, En Chang

Controlling the coupling between different degrees of freedom in many-body systems is a powerful technique for engineering novel phases of matter. We create a bilayer system of two-dimensional (2D) ultracold Bose gases and demonstrate the controlled generation of bulk coherence through tunable interlayer Josephson coupling. We probe the resulting correlation

Renze Lou, Hanzi Xu, Sijia Wang, Jiangshu Du

Numerous studies have assessed the proficiency of AI systems, particularly large language models (LLMs), in facilitating everyday tasks such as email writing, question answering, and creative content generation. However, researchers face unique challenges and opportunities in leveraging LLMs for their own work, such as brainstorming research ideas, designing

Guangqi Jiang, Yifei Sun, Tao Huang, Huanyu Li

The pre-training of visual representations has enhanced the efficiency of robot learning. Due to the lack of large-scale in-domain robotic datasets, prior works utilize in-the-wild human videos to pre-train robotic visual representation. Despite their promising results, representations from human videos are inevitably subject to distribution shifts and lack

Dawei Gao, Yu Yuan, Nicolas Brodusch, Raynald Gauvin

This manuscript presents a comparative analysis of two software packages, MC X-ray and PENELOPE, focusing on their accuracy and efficiency in simulating k-ratios for binary compounds and comparing their spectra with experimental data for pure elements and compounds. Based on the Pouchou database, MC X-ray slightly outperforms PENELOPE in k-ratio calculations

Dylan Gaines, Keith Vertanen

Text input on mobile devices without physical keys can be challenging for people who are blind or low-vision. We interview 12 blind adults about their experiences with current mobile text input to provide insights into what sorts of interface improvements may be the most beneficial. We identify three primary themes that were experiences or opinions shared by

Seetharam Killivalavan, Durairaj Thenmozhi

This paper explores a novel method for enhancing binary classification models that assess code comment quality, leveraging Generative Artificial Intelligence to elevate model performance. By integrating 1,437 newly generated code-comment pairs, labeled as "Useful" or "Not Useful" and sourced from various GitHub repositories, into an existing C-language datas

Taiwo A. Adebiyi, Bach Do, Ruda Zhang

Bayesian optimization devolves the global optimization of a costly objective function to the global optimization of a sequence of acquisition functions. This inner-loop optimization can be catastrophically difficult if it involves posterior sample paths, especially in higher dimensions. We introduce an efficient global optimization strategy for posterior sam

O. Ogulcan Tuncer, I. Yurdusen

We initiate a research program for the systematic investigation of quantum superintegrable systems involving the interaction of two non-relativistic particles with spin $1/2$ moving in the three-dimensional Euclidean space. In this paper, we focus specifically on such superintegrable systems that allow additional scalar integrals of motion, linear in the mom

Nicole K. Guittari, Miguel E. Wimbish, Patricia K. Rivlin, Mark A. Hinton

The promise of large-scale, high-resolution datasets from Electron Microscopy (EM) and X-ray Microtomography (XRM) lies in their ability to reveal neural structures and synaptic connectivity, which is critical for understanding the brain. Effectively managing these complex and rapidly increasing datasets will enable new scientific insights, facilitate queryi

Naren Sengodan

Breast cancer histopathology image classification is critical for early detection and improved patient outcomes. 1 This study introduces a novel approach leveraging EfficientNetV2 models, to improve feature extraction and focus on relevant tissue regions. The proposed models were evaluated on the BreakHis dataset across multiple magnification scales (40X, 10

Thomas Schmied, Thomas Adler, Vihang Patil, Maximilian Beck

In recent years, there has been a trend in the field of Reinforcement Learning (RL) towards large action models trained offline on large-scale datasets via sequence modeling. Existing models are primarily based on the Transformer architecture, which result in powerful agents. However, due to slow inference times, Transformer-based approaches are impractical

Mikita Balesni, Marius Hobbhahn, David Lindner, Alexander Meinke

We sketch how developers of frontier AI systems could construct a structured rationale -- a 'safety case' -- that an AI system is unlikely to cause catastrophic outcomes through scheming. Scheming is a potential threat model where AI systems could pursue misaligned goals covertly, hiding their true capabilities and objectives. In this report, we propose thre

C. M. Raiteri, M. Villata, M. I. Carnerero, S. O. Kurtanidze

Blazars are beamed active galactic nuclei known for their strong multi-wavelength variability on timescales from years down to minutes. We aim to investigate the suitability of the twisting jet model presented in previous works to explain the multi-wavelength behaviour of BL Lacertae, the prototype of one of the blazar classes. According to this model, the j

Can Chen, Jun-Kun Wang

Developing algorithms to differentiate between machine-generated texts and human-written texts has garnered substantial attention in recent years. Existing methods in this direction typically concern an offline setting where a dataset containing a mix of real and machine-generated texts is given upfront, and the task is to determine whether each sample in th

Kai Wang, Fei Yang, Bogdan Raducanu, Joost van de Weijer

With the advent of large pre-trained vision-language models such as CLIP, prompt learning methods aim to enhance the transferability of the CLIP model. They learn the prompt given few samples from the downstream task given the specific class names as prior knowledge, which we term as semantic-aware classification. However, in many realistic scenarios, we onl

Xinyu Zhao, Fangcong Yin, Greg Durrett

Long-context LLMs are increasingly in demand for applications such as retrieval-augmented generation. To defray the cost of pretraining LLMs over long contexts, recent work takes an approach of synthetic context extension: fine-tuning LLMs with synthetically generated long-context data in a post-training stage. However, it remains unclear how and why this sy

Paola Cascante-Bonilla, Yu Hou, Yang Trista Cao, Hal Daumé

Compositional reasoning in Vision-Language Models (VLMs) remains challenging as these models often struggle to relate objects, attributes, and spatial relationships. Recent methods aim to address these limitations by relying on the semantics of the textual description, using Large Language Models (LLMs) to break them down into subsets of questions and answer

Minghao Ning, Ahmad Reza Alghooneh, Chen Sun, Ruihe Zhang

In this paper, we propose an accurate and robust perception module for Autonomous Vehicles (AVs) for drivable space extraction. Perception is crucial in autonomous driving, where many deep learning-based methods, while accurate on benchmark datasets, fail to generalize effectively, especially in diverse and unpredictable environments. Our work introduces a r

Bo Jiang, Shaoyu Chen, Bencheng Liao, Xingyu Zhang

End-to-end autonomous driving demonstrates strong planning capabilities with large-scale data but still struggles in complex, rare scenarios due to limited commonsense. In contrast, Large Vision-Language Models (LVLMs) excel in scene understanding and reasoning. The path forward lies in merging the strengths of both approaches. Previous methods using LVLMs t

Seongmin Lee, Ali Payani, Duen Horng Chau

Modern deep learning models often make predictions by focusing on irrelevant areas, leading to biased performance and limited generalization. Existing methods aimed at rectifying model attention require explicit labels for irrelevant areas or complex pixel-wise ground truth attention maps. We present CRAYON (Correcting Reasoning with Annotations of Yes Or No

Karthik Prakhya, Tolga Birdal, Alp Yurtsever

Solving non-convex, NP-hard optimization problems is crucial for training machine learning models, including neural networks. However, non-convexity often leads to black-box machine learning models with unclear inner workings. While convex formulations have been used for verifying neural network robustness, their application to training neural networks remai

Benjamin Brück, Sam Hughes, Dawid Kielak, Piotr Mizerka

We construct explicit finite-dimensional orthogonal representations $\pi_N$ of $\operatorname{SL}_{N}(\mathbb{Z})$ for $N \in \{3,4\}$ all of whose invariant vectors are trivial, and such that $H^{N - 1}(\operatorname{SL}_{N}(\mathbb{Z}),\pi_N)$ is non-trivial. This implies that for $N$ as above, the group $\operatorname{SL}_{N}(\mathbb{Z})$ does not have pr

Hongze Wang, Jiaxu Xing, Nico Messikommer, Davide Scaramuzza

Reinforcement learning (RL) has achieved outstanding success in complex robot control tasks, such as drone racing, where the RL agents have outperformed human champions in a known racing track. However, these agents fail in unseen track configurations, always requiring complete retraining when presented with new track layouts. This work aims to develop RL ag

Yifan Sun, Yuhang Li, Yue Zhang, Yuchen Jin

The ever-increasing size of open-source Large Language Models (LLMs) renders local deployment impractical for individual users. Decentralized computing has emerged as a cost-effective solution, allowing individuals and small companies to perform LLM inference for users using surplus computational power. However, a computing provider may stealthily substitute

Haomeng Zhang, Chiao-An Yang, Raymond A. Yeh

Multi-object 3D Grounding involves locating 3D boxes based on a given query phrase from a point cloud. It is a challenging and significant task with numerous applications in visual understanding, human-computer interaction, and robotics. To tackle this challenge, we introduce D-LISA, a two-stage approach incorporating three innovations. First, a dynamic visi

Youness Lamzouri, Kunjakanan Nath

For a primitive Dirichlet character $χ\pmod q$ we let \[M(χ):= \frac{1}{\sqrt{q}}\max_{1\leq t \leq q} \Big|\sum_{n \leq t} χ(n) \Big|.\] In this paper, we investigate the distribution of $M(χ)$, as $χ$ ranges over primitive cubic characters $χ\pmod q$ with $(q,3)=1$ and $q\leq Q$. Our first result gives an estimate for the proportion of such characters for

Yihe Deng, Paul Mineiro

Mathematical reasoning is a crucial capability for Large Language Models (LLMs), yet generating detailed and accurate reasoning traces remains a significant challenge. This paper introduces a novel approach to produce high-quality reasoning traces for LLM fine-tuning using online learning \textbf{Flows}. Our method employs an incremental output production Fl

Harish Karthikeyan, Antigoni Polychroniadou

Our work aims to minimize interaction in secure computation due to the high cost and challenges associated with communication rounds, particularly in scenarios with many clients. In this work, we revisit the problem of secure aggregation in the single-server setting where a single evaluation server can securely aggregate client-held individual inputs. Our ke

Ish Gupta, Koustav Chandra, B. S. Sathyaprakash

Next-generation gravitational-wave observatories are expected to detect over a thousand compact binary coalescence signals daily, with some lasting from minutes to hours. Consequently, multiple signals will overlap in the time-frequency plane, generating a "foreground noise" that predominantly affects the low-frequency range, where binary neutron star inspir

Amiran Gogatishvili, Tuğçe Ünver

The main objective of this paper is to provide a comprehensive demonstration of recent results regarding the structures of the weighted Ces\`aro and Copson function spaces. These spaces' definitions involve local and global weighted Lebesgue norms; in other words, the norms of these spaces are generated by positive sublinear operators and by weighted Lebesgu

Gabriel Wallin, Yunxiao Chen, Yi-Hsuan Lee, Xiaoou Li

Educational assessments are valuable tools for measuring student knowledge and skills, but their validity can be compromised when test takers exhibit changes in response behavior due to factors such as time pressure. To address this issue, we introduce a novel latent factor model with change-points for item response data, designed to detect and account for i

Souraja Kundu, Saket Singh, Yuji Iwahori

Generating music from images can enhance various applications, including background music for photo slideshows, social media experiences, and video creation. This paper presents an emotion-guided image-to-music generation framework that leverages the Valence-Arousal (VA) emotional space to produce music that aligns with the emotional tone of a given image. U

Yuhao Zhao, Oded Zilberberg, Antonio Štrkalj

Inverted-band $pn$ junctions in two-dimensional materials offer a promising platform for electron optics in condensed matter, as they allow to manipulate and guide electron beams without the need for spatial confinement. In this work, we propose the realization of an Aharonov-Bohm (AB) interferometer using such $pn$ junctions. We observe AB oscillations in n

João V. P. e Silva

This article focuses on the study of the group of units of incidence rings, which is a class of infinite matrix groups indexed by ordered sets, on a topological perspective. We first show when these groups can inherit the topological structure from the incidence rings. It is later shown that infinite matrix groups of topological fields can be used to build s

Quoc Tran-Dinh, Trang H. Tran, Lam M. Nguyen

This paper aims at developing novel shuffling gradient-based methods for tackling two classes of minimax problems: nonconvex-linear and nonconvex-strongly concave settings. The first algorithm addresses the nonconvex-linear minimax model and achieves the state-of-the-art oracle complexity typically observed in nonconvex optimization. It also employs a new sh

Angelica Chen, Samuel D. Stanton, Frances Ding, Robert G. Alberstein

Although large language models (LLMs) have shown promise in biomolecule optimization problems, they incur heavy computational costs and struggle to satisfy precise constraints. On the other hand, specialized solvers like LaMBO-2 offer efficiency and fine-grained control but require more domain expertise. Comparing these approaches is challenging due to expen

Mohammad Setak, Pooria Madani

Recent advancements in Large Language Models (LLMs) have significantly improved their capabilities in natural language processing and code synthesis, enabling more complex applications across different fields. This paper explores the application of LLMs in the context of code mutation, a process where the structure of program code is altered without changing

Chirag Modi, Diana Cai, Lawrence K. Saul

Black-box variational inference (BBVI) scales poorly to high-dimensional problems when it is used to estimate a multivariate Gaussian approximation with a full covariance matrix. In this paper, we extend the batch-and-match (BaM) framework for score-based BBVI to problems where it is prohibitively expensive to store such covariance matrices, let alone to est

Xuan Hu, Qijun Chen, Nicholas H. Luo, Richy J. Zheng

Peridynamics is a non-local continuum mechanics theory that offers unique advantages for modeling problems involving discontinuities and complex deformations. Within the peridynamic framework, various formulations exist, among which the material correspondence formulation stands out for its ability to directly incorporate traditional continuum material model

Nicholas A. Corbin, Boris Kramer

We consider the optimal regulation problem for nonlinear control-affine dynamical systems. Whereas the linear-quadratic regulator (LQR) considers optimal control of a linear system with quadratic cost function, we study polynomial systems with polynomial cost functions; we call this problem the polynomial-polynomial regulator (PPR). The resulting polynomial

Nikolas A. Wittek, Leor Barack, Harald P. Pfeiffer, Adam Pound

Worldtube excision is a method of reducing computational burden in Numerical Relativity simulations of binary black holes in situations where there is a good analytical model of the geometry around (one or both of) the objects. Two such scenarios of relevance in gravitational-wave astronomy are (1) the case of mass-disparate systems, and (2) the early inspir

Aditya Johri, Ashish Hingle, Johannes Schleiss

In this research paper we examine undergraduate students' use of and perceptions of generative AI (GenAI). Students are early adopters of the technology, utilizing it in atypical ways and forming a range of perceptions and aspirations about it. To understand where and how students are using these tools and how they view them, we present findings from an open

Yiqi Zhong, Luming Liang, Bohan Tang, Ilya Zharkov

We introduce motion graph, a novel approach to the video prediction problem, which predicts future video frames from limited past data. The motion graph transforms patches of video frames into interconnected graph nodes, to comprehensively describe the spatial-temporal relationships among them. This representation overcomes the limitations of existing motion

Noah Lordi, Maedee Trank-Greene, Akira Kyle, Joshua Combes

Permutation puzzles, such as the Rubik's Cube and the 15 puzzle, are enjoyed by the general public and mathematicians alike. Here we introduce quantum versions of permutation puzzles where the pieces of the puzzles are replaced with indistinguishable quantum particles. The moves in the puzzle are achieved by swapping or permuting the particles. We show that

Davood B. Dar, Neepa T. Maitra

A striking example of the need to accurately capture states of double-excitation character in molecules is seen in predicting photo-induced dynamics in small polyenes. Due to the coupling of electronic and nuclear motions,the dark 2$^1$Ag state, known to have double-excitation character, can be reached after an initial photo-excitation to the bright $^1$Bu s

Daniel Defays

Applying the word2vec technique, commonly used in language modeling, to melodies, where notes are treated as words in sentences, enables the capture of pitch information. This study examines two datasets: 20 children's songs and an excerpt from a Bach sonata. The semantic space for defining the embeddings is of very small dimension, specifically 2. Notes are

Md. Ahsan Ayub, Subhabrata Majumdar

Large Language Models (LLMs) are seeing significant adoption in every type of organization due to their exceptional generative capabilities. However, LLMs are found to be vulnerable to various adversarial attacks, particularly prompt injection attacks, which trick them into producing harmful or inappropriate content. Adversaries execute such attacks by craft

Yuanxi Wang, Zuowen Wang, Shih-Chii Liu

This paper presents an efficient deep learning solution for decoding motor movements from neural recordings in non-human primates. An Autoencoder Gated Recurrent Unit (AEGRU) model was adopted as the model architecture for this task. The autoencoder is only used during the training stage to achieve better generalization. Together with the preprocessing techn

Renzhe Yu, Zhen Xu, Sky CH-Wang, Richard Arum

The universal availability of ChatGPT and other similar tools since late 2022 has prompted tremendous public excitement and experimental effort about the potential of large language models (LLMs) to improve learning experience and outcomes, especially for learners from disadvantaged backgrounds. However, little research has systematically examined the real-w

Areej Ali, Aayushi Hingle Collier, Umama Dewan, Nora McDonald

Since the release of ChatGPT in 2022, Generative AI (GenAI) is increasingly being used in higher education computing classrooms across the United States. While scholars have looked at overall institutional guidance for the use of GenAI and reports have documented the response from schools in the form of broad guidance to instructors, we do not know what poli

Nan Cai, Pia Bideau

Event cameras provide a natural and data efficient representation of visual information, motivating novel computational strategies towards extracting visual information. Inspired by the biological vision system, we propose a behavior driven approach for object-wise distance estimation from event camera data. This behavior-driven method mimics how biological

M. Bernades, F. Capuano, F. Duchaine, L. Jofre

A posteriori analysis based upon a recently proposed non-dissipative large-eddy simulation framework for transcritical wall-bounded turbulence has been carried out. Due to the complexities arisen in such flows, the discretization requires kinetic-energy- and pressure-equilibrium-preservation schemes to yield stable and non-dissipative scale-resolving simulat

Yuanheng Wang, Diptarka Hait, Pablo A. Unzueta, Juncheng Harry Zhang

Efficient hybrid DFT simulations of solid state materials would be extremely beneficial for computational chemistry and materials science, but is presently bottlenecked by difficulties in computing Hartree-Fock (HF) exchange with plane wave orbital bases. We present a GPU-accelerated, Gaussian orbital based integral algorithm for systems with periodic bounda

Dahlia R. Klein, Uri Zondiner, Amit Keren, John Birkbeck

Electrons in solids owe their properties to the periodic potential landscapes they experience. The advent of moir\'e lattices has revolutionized our ability to engineer such landscapes on nanometer scales, leading to numerous groundbreaking discoveries. Despite this progress, direct imaging of these electrostatic potential landscapes remains elusive. In this

Samuel J. Boos, Luc Dessart, Ken J. Shen, Dean M. Townsley

Many promising explosion models for the elusive origin of Type Ia supernovae (SNe Ia) ultimately fail to completely reproduce a number of observed properties of these events. One limiting factor for many of these models is the use of the local thermodynamic equilibrium (LTE) assumption in the calculation of their synthetic observables, which has been shown t

Mario Bonk, Janne Junnila, Steffen Rohde, Yilin Wang

In this paper we consider Jordan curves on the Riemann sphere passing through $n \ge 3$ given points. We show that in each relative isotopy class of such curves, there exists a unique curve that minimizes the Loewner energy. These curves have the property that each arc between two consecutive points is a hyperbolic geodesic in the domain bounded by the other

Dorsaf Sallami, Esma Aïmeur

The widespread and diverse online media platforms and other internet-driven communication technologies have presented significant challenges in defining the boundaries of freedom of expression. Consequently, the internet has been transformed into a potential cyber weapon. Within this evolving landscape, two particularly hazardous phenomena have emerged: fake

Antonio M. Coutinho, Anirban Karan, Víctor Miralles, Antonio Pich

Two-Higgs-doublet models come with an augmented parameter space which allows them to possibly solve some of the shortcomings of the Standard Model, and opens the window to a plethora of new phenomena to be discovered. The introduction of scalar-mediated tree-level flavour-changing neutral currents may be tackled with the imposition of extra symmetries on the

Amabile Tatone, Filippo Recrosi, Giuseppe Tomassetti

We give a description of cell diffusion in a soft tissue, paying special attention to the coupling of force, matter, and microforce balance laws through a suitable dissipation principle. To this end, we cast our framework into a multi-level schematics, comprising both kinematics and kinetics, based on a characterization of the free energy. We lay down first

Chengke Zou, Xingang Guo, Rui Yang, Junyu Zhang

The rapid advancements in Vision-Language Models (VLMs) have shown great potential in tackling mathematical reasoning tasks that involve visual context. Unlike humans who can reliably apply solution steps to similar problems with minor modifications, we found that SOTA VLMs like GPT-4o can consistently fail in these scenarios, revealing limitations in their

J. McCullough, A. Amon, E. Legnani, D. Gruen

Modeling the intrinsic alignment (IA) of galaxies poses a challenge to weak lensing analyses. The Dark Energy Survey is expected to be less impacted by IA when limited to blue, star-forming galaxies. The cosmological parameter constraints from this blue cosmic shear sample are stable to IA model choice, unlike passive galaxies in the full DES Y3 sample, the

Davide Berghi, Philip J. B. Jackson

This report describes our systems submitted for the DCASE2024 Task 3 challenge: Audio and Audiovisual Sound Event Localization and Detection with Source Distance Estimation (Track B). Our main model is based on the audio-visual (AV) Conformer, which processes video and audio embeddings extracted with ResNet50 and with an audio encoder pre-trained on SELD, re

Adolfo S. Carvalho, Lynne A. Hillenbrand

The accretion luminosity of an FU Ori disk is a fundamental system parameter, but a challenging one to estimate for all but the most well-studied systems. FU Ori objects are dynamically evolving accretion disks, especially close in time to the outburst epoch. They have a complex multi-temperature disk structure that results in distinctly shaped, broad SEDs.

Nate Gillman, Daksh Aggarwal, Michael Freeman, Saurabh Singh

As the quality of large language models has improved, there has been increased interest in using them to model non-linguistic tokens. For example, the Decision Transformer recasts agentic decision making as a sequence modeling problem, using a decoder-only LLM to model the distribution over the discrete action space for an Atari agent. However, when adapting

D. V. Lomasov, P. N. Vabishchevich

The problems of numerical modeling of viscous incompressible fluid flows are widely considered in computational fluid dynamics. Stationary solutions of boundary value problems for the Navier-Stokes equations exist at large Reynolds numbers, but they are unstable and lead to transient or turbulent unsteady regimes. In addition, the solution of the boundary va

Weiming Feng, Ce Jin

We revisit the classic #Knapsack problem, which asks to count the Boolean points $(x_1,\dots,x_n)\in\{0,1\}^n$ in a given half-space $\sum_{i=1}^nW_ix_i\le T$. This #P-complete problem admits $(1\pm\epsilon)$-approximation. Before this work, [Dyer, STOC 2003]'s $\tilde{O}(n^{2.5}+n^2{\epsilon^{-2}})$-time randomized approximation scheme remains the fastest k

Víctor Hernández-Santamaría, Subrata Majumdar, Luz de Teresa

In this paper, we address the exponential stabilization of the linearized FitzHugh-Nagumo system using an event-triggered boundary control strategy. Employing the backstepping method, we derive a feedback control law that updates based on specific triggering rules while ensuring the exponential stability of the closed-loop system. We establish the well-posed

Amin Ranem, John Kalkhof, Anirban Mukhopadhyay

Medical image registration is a critical process that aligns various patient scans, facilitating tasks like diagnosis, surgical planning, and tracking. Traditional optimization based methods are slow, prompting the use of Deep Learning (DL) techniques, such as VoxelMorph and Transformer-based strategies, for faster results. However, these DL methods often im

Jacob L. Block, Sundararajan Srinivasan, Liam Collins, Aryan Mokhtari

The power of foundation models (FMs) lies in their capacity to learn highly expressive representations that can be adapted to a broad spectrum of tasks. However, these pretrained models require additional training stages to become effective for downstream applications. In the multi-task setting, prior works have shown empirically that specific meta-learning

Ioannis D. Gialamas, Kyriakos Tamvakis

In the context of metric-affine gravity theories, where the metric and connection are independent, we examine actions involving quadratic terms in the Ricci scalar curvature and the Holst invariant. These actions are non-minimally coupled to a scalar field. We explore the behavior of the corresponding effective metric theory, which includes an extra dynamic

Mariam Musavi, Emmanuel Irabor, Abhijit Das, Eduard Alarcon

Next-generation artificial intelligence (AI) workloads are posing challenges of scalability and robustness in terms of execution time due to their intrinsic evolving data-intensive characteristics. In this paper, we aim to analyse the potential bottlenecks caused due to data movement characteristics of AI workloads on scale-out accelerator architectures comp

Peng Fu, Kohei Kishida, Neil J. Ross, Peter Selinger

The quantum programming language Quipper supports circuit operations such as reversing and controlling certain quantum circuits. Additionally, Quipper provides a function called with-computed, which can be used to program circuits of the form g; f; g-dagger. The latter is a common pattern in quantum circuit design. One benefit of using with-computed, as oppo

Peter J. Cameron

This paper makes some preliminary observations towards an extension of current work on graphs defined on groups to simplicial complexes. I define a variety of simplicial complexes on a group which are preserved by automorphisms of the group, and in many cases have a relation to familiar graphs on the group. The ones which seem to reach deepest into the graph

Corey Brooke, Sarah Frei, Lisa Marquand

We give several examples of pairs of non-isomorphic cubic fourfolds whose Fano varieties of lines are birationally equivalent (and in one example isomorphic). Two of our examples, which are special families of conjecturally irrational cubics in $\calC_{12}$, provide new evidence for the conjecture that Fourier-Mukai partners are birationally equivalent. We e

Patricia Pauli, Ruigang Wang, Ian Manchester, Frank Allgöwer

We propose a novel layer-wise parameterization for convolutional neural networks (CNNs) that includes built-in robustness guarantees by enforcing a prescribed Lipschitz bound. Each layer in our parameterization is designed to satisfy a linear matrix inequality (LMI), which in turn implies dissipativity with respect to a specific supply rate. Collectively, th

Farima Fatahi Bayat, Lechen Zhang, Sheza Munir, Lu Wang

The rapid adoption of language models (LMs) across diverse applications has raised concerns about their factuality, i.e., their consistency with real-world facts. We first present VERIFY (Verification and Evidence RetrIeval for FactualitY evaluation), a pipeline to evaluate LMs' factuality in real-world user interactions. VERIFY considers the verifiability o

Hongyi Xu

Multivariate time series anomaly detection technology plays an important role in many fields including aerospace, water treatment, cloud service providers, etc. Excellent anomaly detection models can greatly improve work efficiency and avoid major economic losses. However, with the development of technology, the increasing size and complexity of data, and th

Deepak Gupta, Bart Cleuren

The cost of stochastic resetting is considered within the context of a discrete random walk model. In addition to standard stochastic resetting, for which a reset occurs with a certain probability after \emph{each} step, we introduce a novel resetting protocol which we dubbed {\it dynamic resetting}. This protocol entails an additional dynamic constraint rel

Haitz Sáez de Ocáriz Borde, Artem Lukoianov, Anastasis Kratsios, Michael Bronstein

We propose Scalable Message Passing Neural Networks (SMPNNs) and demonstrate that, by integrating standard convolutional message passing into a Pre-Layer Normalization Transformer-style block instead of attention, we can produce high-performing deep message-passing-based Graph Neural Networks (GNNs). This modification yields results competitive with the stat

Chansup Byun, Albert Reuther, LaToya Anderson, William Arcand

There is a tremendous amount of interest in AI/ML technologies due to the proliferation of generative AI applications such as ChatGPT. This trend has significantly increased demand on GPUs, which are the workhorses for training AI models. Due to the high costs of GPUs and lacking supply, it has become of interest to optimize GPU usage in HPC centers. MIT Lin

Mohammad Anis, Srinivas R. Geedipally, Dominique Lord

Pedestrian safety remains a pressing concern near bus stops along urban transit, where frequent pedestrian-vehicle interactions occur. While prior research has primarily focused on intersections and midblock locations, bus stops have often been treated as secondary contributors rather than as distinct sites requiring targeted safety assessments. This has lef

Mohammad Hossein Rahimi Abkenar, Ahmad Mohamadnejad, Reza Sepahvand

We investigate a beyond Standard Model (SM) featuring five new fields. Four fields encompassing three distinct spin states - scalar ($ S $), spinor ($ \psi^{1,2} $), and vector ($ V_{\mu} $) - together form the multi-component dark matter (DM), while the fifth (scalar) field ($ \phi $) carries a unit charge under a dark $ U_{D}(1) $ gauge symmetry, enabling

Gilles Abramovici

Topologically protected states can be found in physical systems, that show singularities in some energy contour diagram. These singularities can be characterized by winding numbers, defined on a classification surface, which maps physical state parameters. We have found a classification surface, which applies for three-band hamiltonian systems in the same wa

Evgeny P. Kurbatov, Vasily Belokurov, Sergey Koposov, Andrey Kravtsov

The innermost portions of the Milky Way's stellar halo have avoided scrutiny until recently. The lack of wide-area survey data, made it difficult to reconstruct an uninterrupted view of the density distribution of the metal-poor stars inside the Solar radius. In this study, we utilize red giant branch (RGB) stars from Gaia, with metallicities estimated using

Rishabh Jain, Vivek M. Bhasi, Adwait Jog, Anand Sivasubramaniam

Personalized recommendation is a ubiquitous application on the internet, with many industries and hyperscalers extensively leveraging Deep Learning Recommendation Models (DLRMs) for their personalization needs (like ad serving or movie suggestions). With growing model and dataset sizes pushing computation and memory requirements, GPUs are being increasingly

Bryon Aragam, Ruiyi Yang

Multivariate distributions often carry latent structures that are difficult to identify and estimate, and which better reflect the data generating mechanism than extrinsic structures exhibited simply by the raw data. In this paper, we propose a model-free approach for estimating such latent structures whenever they are present, without assuming they exist a

Gabriele Gemmi, Michele Polese, Tommaso Melodia, Leonardo Maccari

Next-generation wireless networks target high network availability, ubiquitous coverage, and extremely high data rates for mobile users. This requires exploring new frequency bands, e.g., mmWaves, moving toward ultra-dense deployments in urban locations, and providing ad hoc, resilient connectivity in rural scenarios. The design of the backhaul network plays

Sylwia Cichacz

We provide a summary of research on disjoint zero-sum subsets in finite Abelian groups, which is a branch of additive group theory and combinatorial number theory. An orthomorphism of a group $\Gamma$ is defined as a bijection $\varphi$ $\Gamma$ such that the mapping $g \mapsto g^{-1}\varphi(g)$ is also bijective. In 1981, Friedlander, Gordon, and Tannenbaum

Pulkit Gopalani, Ekdeep Singh Lubana, Wei Hu

Recent analysis on the training dynamics of Transformers has unveiled an interesting characteristic: the training loss plateaus for a significant number of training steps, and then suddenly (and sharply) drops to near--optimal values. To understand this phenomenon in depth, we formulate the low-rank matrix completion problem as a masked language modeling (ML

P. S. Krstic, A. Maan. R. Majeski, B. E. Koel

We investigate how growth of an oxide film will influence the deuterium recycling properties of Li on plasma facing surfaces. Lithium films on the walls or plasma-facing material surfaces of a fusion vacuum vessel improves plasma performance in part by removing residual impurity atoms from the plasma. Oxygen atoms, from residual water vapor or eroded oxide s

Yuan Luo, Dmitriy Morozov, Luis Scoccola

The Betti tables of a multigraded module encode the grades at which there is an algebraic change in the module. Multigraded modules show up in many areas of pure and applied mathematics, and in particular in topological data analysis, where they are known as persistence modules, and where their Betti tables describe the places at which the homology of filter

Elia Fusi, Federico Giusti

Given a compact Chern-Ricci flat balanced orbifold, we show that its blow-up at a finite family of smooth points admits constant Chern scalar curvature balanced metrics, extending Arezzo-Pacard's construction to the balanced setting. Moreover, if the orbifold has isolated singularities and admits crepant resolutions, we show that they always carry Chern-Ricc

Yuxuan Chen, Mingwei Liu, Guangsheng Ou, Anji Li

Code search is essential for code reuse, allowing developers to efficiently locate relevant code snippets. The advent of powerful decoder-only Large Language Models (LLMs) has revolutionized many code intelligence tasks. However, their effectiveness for the retrieval-based task of code search, particularly compared to established encoder-based models, remain

Rakesh R. Menon, Shashank Srivastava

Despite their high predictive accuracies, current machine learning systems often exhibit systematic biases stemming from annotation artifacts or insufficient support for certain classes in the dataset. Recent work proposes automatic methods for identifying and explaining systematic biases using keywords. We introduce DISCERN, a framework for interpreting sys

Rupert L. Frank, Simon Larson

We prove two-term spectral asymptotics for the Riesz means of the eigenvalues of the Laplacian on a Lipschitz domain with Robin boundary conditions. The second term is the same as in the case of Neumann boundary conditions. This is valid for Riesz means of arbitrary positive order. For orders at least one and under additional assumptions on the function dete

Aleksandros Sobczyk

Optimizing data movements during program executions is essential for achieving high performance in modern computing systems. This has been classically modeled with the Red-Blue Pebble Game and its variants. In existing models, it is typically assumed that the number of red pebbles, i.e., the size of the fast memory, is larger than the maximum in-degree in th