October 2024 arXiv papers — page 113

Qianggang Ding, Haochen Shi, Jiadong Guo, Bang Liu

The integration of Artificial Intelligence (AI) in the financial domain has opened new avenues for quantitative trading, particularly through the use of Large Language Models (LLMs). However, the challenge of effectively synthesizing insights from diverse data sources and integrating both structured and unstructured data persists. This paper presents TradeEx

Zhaocheng Zhu

Reasoning, the ability to logically draw conclusions from existing knowledge, is a hallmark of human. Together with perception, they constitute the two major themes of artificial intelligence. While deep learning has pushed the limit of perception beyond human-level performance, the progress in reasoning domains is way behind. One fundamental reason is that

Elia Mazzucchelli

Some WZW models on affine Lie superalgebras at critical level describe string theory on AdS backgrounds at critical values of the string tension. This is the case of $\mathfrak{psu}(1,1|2)_1$ for ${\rm AdS}_3 \times {\rm S}^3$ and potentially of $\mathfrak{u}(2|2)_1$ (or related algebras) for ${\rm AdS}_5 \times {\rm S}^5$. Many interesting features of these

Fawaz Sammani, Nikos Deligiannis

Contrastive Language-Image Pretraining (CLIP) performs zero-shot image classification by mapping images and textual class representation into a shared embedding space, then retrieving the class closest to the image. This work provides a new approach for interpreting CLIP models for image classification from the lens of mutual knowledge between the two modali

Keshav B. Patel, Wolfgang Bergmeier, Aaron L. Fogelson

Through experimental studies, many details of the pathway of integrin $\alpha_{\rm IIb}\beta_3$ activation by ADP during the platelet aggregation process have been mapped out. ADP binds to two separate G protein coupled receptors on platelet surfaces, leading to alterations in the regulation of the small GTPase RAP1. We seek to (1) gain insights into the rel

Velibor Bojković, Jovana Nikolić, Mladen Zekić

It is clear that every rational surgery on a Hopf link in $3$-sphere is a lens space surgery. In this note we give an explicit computation which lens space is a resulting manifold. The main tool we use is the calculus of continued fractions. As a corollary, we recover the (well known) result on the criterion for when rational surgery on a Hopf link gives the

Faizan Faisal, Umair Yousaf

We present LEGAL-UQA, the first Urdu legal question-answering dataset derived from Pakistan's constitution. This parallel English-Urdu dataset includes 619 question-answer pairs, each with corresponding legal article contexts, addressing the need for domain-specific NLP resources in low-resource languages. We describe the dataset creation process, including

Idan Attias, Steve Hanneke, Arvind Ramaswami

We present novel reductions from sample compression schemes in multiclass classification, regression, and adversarially robust learning settings to binary sample compression schemes. Assuming we have a compression scheme for binary classes of size $f(d_\mathrm{VC})$, where $d_\mathrm{VC}$ is the VC dimension, then we have the following results: (1) If the bi

Sergio Blanes, Fernando Casas, Cesareo Gonzalez, Mechthild Thalhammer

This contribution is dedicated to the exploration of exponential operator splitting methods for the time integration of evolution equations. It entails the review of previous achievements as well as the depiction of novel results. The standard class of splitting methods involving real coefficients is contrasted with an alternative approach that relies on the

Arka Daw, Megan Hong-Thanh Chung, Maria Mahbub, Amir Sadovnik

Machine learning models are known to be vulnerable to adversarial attacks, but traditional attacks have mostly focused on single-modalities. With the rise of large multi-modal models (LMMs) like CLIP, which combine vision and language capabilities, new vulnerabilities have emerged. However, prior work in multimodal targeted attacks aim to completely change t

Leif Hancox-Li, Borhane Blili-Hamelin

ETHICS is probably the most-cited dataset for testing the ethical capabilities of language models. Drawing on moral theory, psychology, and prompt evaluation, we interrogate the validity of the ETHICS benchmark. Adding to prior work, our findings suggest that having a clear understanding of ethics and how it relates to empirical phenomena is key to the valid

Paul Seymour

We give a construction to build all digraphs with the property that every directed cycle has length three.

Rahul Krishna, Rangeet Pan, Raju Pavuluri, Srikanth Tamilselvam

Large Language Models for Code (or code LLMs) are increasingly gaining popularity and capabilities, offering a wide array of functionalities such as code completion, code generation, code summarization, test generation, code translation, and more. To leverage code LLMs to their full potential, developers must provide code-specific contextual information to t

David Farr, Nico Manzonelli, Iain Cruickshank, Kate Starbird

The ability of large language models (LLMs) to perform zero-shot classification makes them viable solutions for data annotation in rapidly evolving domains where quality labeled data is often scarce and costly to obtain. However, the large-scale deployment of LLMs can be prohibitively expensive. This paper introduces an LLM chain ensemble methodology that al

Thijs Bon, Vincent Van Craenenbroeck, Johan Meyers

Integrating floating photovoltaic (FPV) installations into offshore wind farms has been proposed as a major opportunity to scale up offshore renewable energy generation. The interaction between these hybrid wind-solar farms and the atmospheric boundary layer (ABL) is for the first time addressed in the present study. Idealized large-eddy simulations (LES) ar

Nikol Chantzi, Ioannis Mouratidis, Ilias Georgakopoulos-Soares

Zimin words are words that have the same prefix and suffix. They are unavoidable patterns, with all sufficiently large strings encompassing them. Here, we examine for the first time the presence of k-mers not containing any Zimin patterns, defined hereafter as Zimin avoidmers, in the human genome. We report that in the reference human genome all k-mers above

Sicheng Wang, Eugenio Frias-Miranda, Antonio Alvarez Valdivia, Laura H. Blumenschein

Soft robots are known for their ability to perform tasks with great adaptability, enabled by their distributed, non-uniform stiffness and actuation. Bending is the most fundamental motion for soft robot design, but creating robust, and easy-to-fabricate soft bending joint with tunable properties remains an active problem of research. In this work, we demonst

Makram Chahine, Alex Quach, Alaa Maalouf, Tsun-Hsuan Wang

End-to-end learning directly maps sensory inputs to actions, creating highly integrated and efficient policies for complex robotics tasks. However, such models often struggle to generalize beyond their training scenarios, limiting adaptability to new environments, tasks, and concepts. In this work, we investigate the minimal data requirements and architectur

Jeffrey P Hatch, Alan E Rask, Duy-Khoi Dang, Paul M Zimmerman

Incremental full configuration interaction (iFCI) is polynomial-cost approach to the FCI limit of electronic structure. This article introduces the many-body basis set amelioration (MBBSA) method, which is designed to allow iFCI to be applicable to larger atomic orbital basis sets. MBBSA uses a series of inexpensive iFCI calculations to approximate the corre

David Bolin, Vaibhav Mehandiratta, Alexandre B. Simas

The computational cost for inference and prediction of statistical models based on Gaussian processes with Mat\'ern covariance functions scales cubicly with the number of observations, limiting their applicability to large data sets. The cost can be reduced in certain special cases, but there are currently no generally applicable exact methods with linear co

Batuhan K. Karaman, Ishmam Zabir, Alon Benhaim, Vishrav Chaudhary

Achieving both high safety and high usefulness simultaneously in large language models has become a critical challenge in recent years.Models often exhibit unsafe behavior or adopt an overly cautious approach leading to frequent overrefusal of benign prompts, which reduces their usefulness. A major factor underlying these behaviors is how the models are fine

Diego Noja, Francesco Raso Stoia

In this paper we describe the resonances of the singular perturbation of the Laplacian on the half space $\Omega =\mathbb R^3_+$ given by the self-adjoint operator named $\delta$-interaction. We will assume Dirichlet or Neumann boundary conditions on $\partial \Omega$. At variance with the well known case of $\mathbb R^3$, the resonances constitute an infini

Kaveh Eskandari Miandoab, Vasanth Sarathy

Large Language Models (LLMs), despite achieving state-of-the-art results in a number of evaluation tasks, struggle to maintain their performance when logical reasoning is strictly required to correctly infer a prediction. In this work, we propose Argument Generation as a method of forcing models to utilize their reasoning capabilities when other approaches s

Nazanin Fouladgar, Marjan Alirezaie, Kary Främling

Local explanation of machine learning (ML) models has recently received significant attention due to its ability to reduce ambiguities about why the models make specific decisions. Extensive efforts have been invested to address explainability for different data types, particularly images. However, the work on multivariate time series data is limited. A poss

Anthony Opipari, Aravindhan K Krishnan, Shreekant Gayaka, Min Sun

This paper presents a method for generating large-scale datasets to improve class-agnostic video segmentation across robots with different form factors. Specifically, we consider the question of whether video segmentation models trained on generic segmentation data could be more effective for particular robot platforms if robot embodiment is factored into th

Zachary Grey, Nicholas Fisher, Andrew Glaws

Scientists, engineers, biologists, and technology specialists universally leverage image segmentation to extract shape ensembles containing many thousands of curves representing patterns in observations and measurements. These large curve ensembles facilitate inferences about important changes when comparing and contrasting images. We introduce novel pattern

Marcela Ordorica Arango, Anastasia Bizyaeva, Simon A. Levin, Naomi Ehrich Leonard

We present and analyze a mathematical model to study the feedback between behavior and epidemic spread in a population that is actively assessing and reacting to risk of infection. In our model, a population dynamically forms an opinion that reflects its willingness to engage in risky behavior (e.g., not wearing a mask in a crowded area) or reduce it (e.g.,

Mikhail V. Medvedev

The model explaining the spectral "zebra" pattern of the high-frequency interpulse (HFIP) of the Crab pulsar radio emission is proposed. The observed emission bands are diffraction fringes in the spectral domain. The pulsar's own plasma-filled magnetosphere plays a role of a frequency-dependent "diffraction screen". The observed features such as the proporti

Daniel Bennett, Elisa Mekler

Motivation and autonomy are fundamental concepts in Human-Computer Interaction (HCI), yet in User Experience (UX) research they have remained surprisingly peripheral. We draw on Self-Determination Theory (SDT) to analyse autonomous and non-autonomous patterns of motivation in 497 interaction experiences. Using latent profile analysis, we identify 5 distinct

Jonghyeon Nam, Jaeduk Han, Hokeun Kim

The 3-level pulse amplitude modulation (PAM-3) signaling is expected to be widely used in memory interfaces for its greater voltage margins compared to PAM-4. To maximize the benefit of PAM-3, we propose three low-power data encoding algorithms: PAM3-DBI, PAM3-MF, and PAM3-SORT. With the DRAM memory traces from the gem5 computer architecture simulator runnin

Iaroslav Chelombitko, Egor Safronov, Aleksey Komissarov

In the development of Large Language Models (LLMs), considerable attention has been given to the quality of training datasets. However, the role of tokenizers in the LLM training pipeline, particularly for multilingual models, has received less focus. The quality of tokenization can significantly impact a model's ability to handle diverse languages effective

Jesús Alejandro Loera-Ponce, Diego A. Mercado-Ravell, Israel Becerra-Durán, Luis Manuel Valentin-Coronado

In this paper, we address the vision-based autonomous landing problem in complex urban environments using deep neural networks for semantic segmentation and risk assessment. We propose employing the SegFormer, a state-of-the-art visual transformer network, for the semantic segmentation of complex, unstructured urban environments. This approach yields valuabl

Samuel C. Lange, Aristeidis Amvrosiadis, James W. Nightingale, Qiuhan He

We analyze two galaxy-scale strong gravitational lenses, SPT0418-47 and SPT2147-50, using JWST NIRCam imaging across multiple filters. To account for angular complexity in the lens mass distribution, we introduce multipole perturbations with orders $m=1, 3, 4$. Our results show strong evidence for angular mass complexity in SPT2147, with multipole strengths

Miłosz Panfil, Zoran Ristivojevic

We study the rapidity distribution in the Lieb-Liniger model and derive exact relations for its derivatives at the Fermi level. The latter enables us to treat analytically the free energy of the system at low temperatures and arbitrary interactions. We calculated the leading-order correction to the well-known result obtained using conformal field theory. In

Konstantinos Skianis, A. Seza Doğruöz, John Pavlopoulos

Large language models (LLMs) are increasingly used in medical fields. In mental health support, the early identification of linguistic markers associated with mental health conditions can provide valuable support to mental health professionals, and reduce long waiting times for patients. Despite the benefits of LLMs for mental health support, there is limite

Zirui Song, Guangxian Ouyang, Meng Fang, Hongbin Na

Existing household robots have made significant progress in performing routine tasks, such as cleaning floors or delivering objects. However, a key limitation of these robots is their inability to recognize potential problems or dangers in home environments. For example, a child may pick up and ingest medication that has fallen on the floor, posing a serious

Stefan Jaeger

Contemporary machine learning methods will try to approach the Bayes error, as it is the lowest possible error any model can achieve. This paper postulates that any decision is composed of not one but two Bayesian decisions and that decision-making is, therefore, a double-Bayesian process. The paper shows how this duality implies intrinsic uncertainty in dec

Jinzhu Luo, Dingyang Chen, Qi Zhang

Data augmentation creates new data points by transforming the original ones for a reinforcement learning (RL) agent to learn from, which has been shown to be effective for the objective of improving the data efficiency of RL for continuous control. Prior work towards this objective has been largely restricted to perturbation-based data augmentation where new

Costin-Andrei Oncescu, Sanket Purandare, Stratos Idreos, Sham Kakade

While transformers have been at the core of most recent advancements in sequence generative models, their computational cost remains quadratic in sequence length. Several subquadratic architectures have been proposed to address this computational issue. Some of them, including long convolution sequence models (LCSMs), such as Hyena, address this issue at tra

Asaf Ferber, Bryce Frederickson, Dingjia Mao, Liana Yepremyan

In 1972, Kotzig proved that for every even $n$, the complete graph $K_n$ can be decomposed into $\lceil\log_2n\rceil$ edge-disjoint regular bipartite spanning subgraphs, which is best possible. In this paper, we study regular bipartite decompositions of $(n,d,\lambda)$-graphs, where $n$ is an even integer and $d_0\leq d\leq n-1$ for some absolute constant $d

Jaroslav Scheinpflug

In this short note, we study the infinite-dimensional symmetry algebras which appear in holomorphic twists of 4d $\mathcal{N}=1$ supersymmetric quantum field theories. In particular, we investigate whether their representation theory helps us understand the semi-chiral ring of $\frac{1}{4}$-BPS operators. We focus on the supersymmetric analogue of $\phi^4$ t

Haechan Mark Bong, Ricardo de Azambuja, Giovanni Beltrame

Real-time aerial image segmentation plays an important role in the environmental perception of Uncrewed Aerial Vehicles (UAVs). We introduce BlabberSeg, an optimized Vision-Language Model built on CLIPSeg for on-board, real-time processing of aerial images by UAVs. BlabberSeg improves the efficiency of CLIPSeg by reusing prompt and model features, reducing c

Hai Cheng, Salvatore D'Oro, Rajeev Gangula, Sakthivel Velumani

Network slicing allows Telecom Operators (TOs) to support service provisioning with diverse Service Level Agreements (SLAs). The combination of network slicing and Open Radio Access Network (RAN) enables TOs to provide more customized network services and higher commercial benefits. However, in the current Open RAN community, an open-source end-to-end slicin

Richard E. Zeebe, Ilja J. Kocken

Astronomical solutions provide calculated orbital and rotational parameters of solar system bodies based on the dynamics and physics of the solar system. Application of astronomical solutions in the Earth sciences has revolutionized our understanding in at least two areas of active research. (i) The Astronomical (or Milankovic) forcing of climate on time sca

Antonio Alex-Amor, Grigorii Ptitcyn, Nader Engheta

With his formal analysis in 1951, the physicist Pyotr Kapitza demonstrated that an inverted pendulum with an externally vibrating base can be stable in its upper position, thus overcoming the force of gravity. Kapitza's work is an example that an originally unstable system can become stable after a minor perturbation of its properties or initial conditions i

D. Ramsey, B. Malaca, T. T. Simpson, M. Formanek

Laser-driven free-electron lasers (LDFELs) replace magnetostatic undulators with the electromagnetic fields of a laser pulse. Because the undulator period is half the wavelength of the laser pulse, LDFELs can amplify x rays using lower electron energies and over shorter interaction lengths than a traditional free-electron laser. In LDFELs driven by conventio

Anna Sokol, Elizabeth Daly, Michael Hind, David Piorkowski

Large language models (LLMs) are powerful tools capable of handling diverse tasks. Comparing and selecting appropriate LLMs for specific tasks requires systematic evaluation methods, as models exhibit varying capabilities across different domains. However, finding suitable benchmarks is difficult given the many available options. This complexity not only inc

Thomas Hangelbroek, Christian Rieger, Grady B. Wright

We present a general framework, treating Lipschitz domains in Riemannian manifolds, that provides conditions guaranteeing the existence of norming sets and generalized local polynomial reproduction - a powerful tool used in the analysis of various mesh-free methods and a mesh-free method in its own right. As a key application, we prove the existence of smoot

Rudra Murthy, Praveen Venkateswaran, Prince Kumar, Danish Contractor

LLM evaluation benchmarks have traditionally separated the testing of knowledge/reasoning capabilities from instruction following. In this work, we study the interaction between knowledge and instruction following, and observe that LLMs struggle to follow simple answer modifying instructions, and are also distracted by instructions that should have no bearin

Shaoyang Xu, Yongqi Leng, Linhao Yu, Deyi Xiong

As large language models (LLMs) become increasingly accessible in many countries, it is essential to align them to serve pluralistic human values across cultures. However, pluralistic culture alignment in LLMs remain an open problem. In this paper, we propose CultureSPA, a Self-Pluralising Culture Alignment framework that allows LLMs to simultaneously align

B. Sazdović

Our main proposition is that field equations for all spins can be obtained from Casimir eigenvalue equations for Poincare group. We have already confirm that statement for massive scalar, spinor and vector fields in Ref.[1]. In the present article we are going to confirm this statement for massless vector, second and fourth rang tensor fields. In particular

A. R. Livernois, F. I. Aros, E. Vesperini, A. Askar

We present the results of Monte Carlo simulations aimed at exploring the evolution towards energy equipartition of first- (1G) and second-generation (2G) stars in multiple-population globular clusters and how this evolution is affected by the initial differences between the spatial distributions of the two populations. Our results show that these initial dif

Dip Kiran Pradhan Newar, Max Fowler, David H. Smith, Seth Poulsen

Background and Context: Some skills taught in introductory programming courses are categorized into 1) explaining code, 2) arranging lines of code in correct sequence, 3) tracing through the execution of a program, and 4) writing code from scratch. Objective: Knowing if a programming skill is a prerequisite to another would benefit teachers in properly plann

Jugal Garg, Eklavya Sharma

We study fair division of indivisible mixed manna when agents have unequal entitlements, with weighted envy-freeness up to one item (WEF1) as our primary notion of fairness. We identify several shortcomings of existing techniques to achieve WEF1. Hence, we relax WEF1 to weighted envy-freeness up to 1 transfer (WEF1T), and give a polynomial-time algorithm for

Piotr Sowinski, Maria Ganzha

Collaborative mechanisms allow benchmarks to be updated continuously and adjust to the changing requirements and new use cases. This paradigm is employed for example in the field of machine learning, but up until now there were no examples of truly open and collaborative benchmarks for RDF systems. In this demo paper we present the collaboration functionalit

Asaf Ferber, Marcelo Sales, Mason Shurman

A covering of a digraph $D$ by Hamilton cycles is a collection of directed Hamilton cycles (not necessarily edge-disjoint) that together cover all the edges of $D$. We prove that for $1/2 \geq p\geq \frac{\log^{20} n}{n}$, the random digraph $D_{n,p}$ typically admits an optimal Hamilton cycle covering. Specifically, the edges of $D_{n,p}$ can be covered by

Timo Hillmann, Guillaume Dauphinais, Ilan Tzitrin, Michael Vasmer

Photonics provides a viable path to a scalable fault-tolerant quantum computer. The natural framework for this platform is measurement-based quantum computation, where fault-tolerant graph states supersede traditional quantum error-correcting codes. However, the existing formalism for foliation - the construction of fault-tolerant graph states - does not rev

Carlos Gustavo Moreira, Jinghua Xi, Yiwei Zhang

Bandt and Kravchenko \cite{BandtKravchenko2010} proved that if a self-similar set spans $\R^m$, then there is no tangent hyperplane at any point of the set. In particular, this indicates that a smooth planar curve is self-similar if and only if it is a straight line. When restricting curves to graphs of continuous functions, we can show that the graph of a c

Yang Liu, Yaofang Liu, Jinshan Pan, Yuxiang Hui

Most existing super-resolution methods and datasets have been developed to improve the image quality in well-lighted conditions. However, these methods do not work well in real-world low-light conditions as the images captured in such conditions lose most important information and contain significant unknown noises. To solve this problem, we propose a SRRIIE

Milan Curcic

Hydrodynamic modulation of short ocean surface waves by longer ambient waves significantly influences remote sensing, interpretation of in situ wave measurements, and numerical wave forecasting. This paper revisits the wave crest and action conservation laws and derives steady, nonlinear, analytical solutions for the change of short-wave wavenumber, action,

Hannah YoungEun An, Lenhart K. Schubert

Commonsense knowledge is essential for machines to reason about the world. Large language models (LLMs) have demonstrated their ability to perform almost human-like text generation. Despite this success, they fall short as trustworthy intelligent systems, due to the opacity of the basis for their answers and a tendency to confabulate facts when questioned ab

Maria Carvalho, Vinícius Coelho, Luciana Salgado

We derive a necessary and sufficient condition for a homeomorphism with the shadowing property to be topologically transitive: to have an invariant subset $A$, dense in the non-wandering set, where the barycenter property holds. To elucidate its dynamical nature, we compare this condition with other properties known to be sufficient for an Anosov diffeomorph

Ruiqi Li, Siqi Zheng, Xize Cheng, Ziang Zhang

Generating music that aligns with the visual content of a video has been a challenging task, as it requires a deep understanding of visual semantics and involves generating music whose melody, rhythm, and dynamics harmonize with the visual narratives. This paper presents MuVi, a novel framework that effectively addresses these challenges to enhance the cohes

Sangheon Park, Danbinaerin Han, Dasaem Jeong

Pansori is one of the most representative vocal genres of Korean traditional music, which has an elaborated vocal melody line with strong vibrato. Although the music is transmitted orally without any music notation, transcribing pansori music in Western staff notation has been introduced for several purposes, such as documentation of music, education, or res

Lu Pang, Tao Sun, Weimin Lyu, Haibin Ling

Recently, backdoor attack has become an increasing security threat to deep neural networks and drawn the attention of researchers. Backdoor attacks exploit vulnerabilities in third-party pretrained models during the training phase, enabling them to behave normally for clean samples and mispredict for samples with specific triggers. Existing backdoor attacks

Ali Borji

The study conducted by Shumailov et al. (2024) demonstrates that repeatedly training a generative model on synthetic data leads to model collapse. This finding has generated considerable interest and debate, particularly given that current models have nearly exhausted the available data. In this work, we investigate the effects of fitting a distribution (thr

Aayush Agrawal, Aniruddh Sikdar, Rajini Makam, Suresh Sundaram

Underwater mine detection with deep learning suffers from limitations due to the scarcity of real-world data. This scarcity leads to overfitting, where models perform well on training data but poorly on unseen data. This paper proposes a Syn2Real (Synthetic to Real) domain generalization approach using diffusion models to address this challenge. We demonstra

Mingyang Chen, Haoze Sun, Tianpeng Li, Fan Yang

Large Language Models (LLMs) have exhibited significant potential in performing diverse tasks, including the ability to call functions or use external tools to enhance their performance. While current research on function calling by LLMs primarily focuses on single-turn interactions, this paper addresses the overlooked necessity for LLMs to engage in multi-t

Raelyn Marguerite Sullivan, Lukas Tobias Hergt, Douglas Scott

This introductory guide aims to provide insight to new researchers in the field of cosmic microwave background (CMB) map analysis on best practices for several common procedures. We will discuss common map-modifying procedures such as masking, downgrading resolution, the effect of the beam and the pixel window function, and adding white noise. We will explor

Carlos A. R. Herdeiro

Recently, a scalar counterpart of the Schwarzschild-Melvin Universe was reported [arXiv:2410.02851]. We show this solution is a special case of a Schwarzschild black hole/mass in a scalar multipolar Universe, that can be constructed algebraically combining known vacuum solutions. This builds on the generalized Weyl construction for scalar-vacuum, that admits

Phillip Guo, Aaquib Syed, Abhay Sheshadri, Aidan Ewart

Methods for knowledge editing and unlearning in large language models seek to edit or remove undesirable knowledge or capabilities without compromising general language modeling performance. This work investigates how mechanistic interpretability -- which, in part, aims to identify model components (circuits) associated to specific interpretable mechanisms t

Abdul Waheed, Hanin Atwany, Bhiksha Raj, Rita Singh

Understanding how speech foundation models capture non-verbal cues is crucial for improving their interpretability and adaptability across diverse tasks. In our work, we analyze several prominent models such as Whisper, Seamless, Wav2Vec, HuBERT, and Qwen2-Audio focusing on their learned representations in both paralinguistic and non-paralinguistic tasks fro

Orchid Chetia Phukan, Devyani Koshal, Swarup Ranjan Behera, Arun Balaji Buduru

Speech forensic tasks (SFTs), such as automatic speaker recognition (ASR), speech emotion recognition (SER), gender recognition (GR), and age estimation (AE), find use in different security and biometric applications. Previous works have applied various techniques, with recent studies focusing on applying speech foundation models (SFMs) for improved performa

Olga Mazur, Leonid Stefanovich

Within the framework of Landau-Ginzburg theory the kinetics of domain structure formation in ferroelectrics that undergo first order phase transition was investigated under the influence of hydrostatic pressure. It was established that mechanical action increases the tendency of nonequilibrium system to the formation of stable polydomain structure. Numerical

Elise Paradis, Kate Grey, Quinn Madison, Daye Nam

How much does AI assistance impact developer productivity? To date, the software engineering literature has provided a range of answers, targeting a diversity of outcomes: from perceived productivity to speed on task and developer throughput. Our randomized controlled trial with 96 full-time Google software engineers contributes to this literature by sharing

Sheng-Chieh Lin, Yuanyuan Su, Fabio Gastaldello, Nathan Jacobs

Inverse Compton (IC) emission associated with the non-thermal component of the intracluster medium (ICM) has been a long sought phenomenon in cluster physics. Traditional spectral fitting often suffers from the degeneracy between the two-temperature thermal spectrum (2T) and the one-temperature plus IC power-law spectrum (1T+IC). We present a semi-supervised

Anugrah Jo Joshy, John T. Hwang

Recent advances in computing hardware and modeling software have given rise to new applications for numerical optimization. These new applications occasionally uncover bottlenecks in existing optimization algorithms and necessitate further specialization of the algorithms. However, such specialization requires expert knowledge of the underlying mathematical

Jintao Ren, Kim Hochreuter, Mathis Ersted Rasmussen, Jesper Folsted Kallehauge

Radiation therapy (RT) is a vital part of treatment for head and neck cancer, where accurate segmentation of gross tumor volume (GTV) is essential for effective treatment planning. This study investigates the use of pre-RT tumor regions and local gradient maps to enhance mid-RT tumor segmentation for head and neck cancer in MRI-guided adaptive radiotherapy.

Jintao Ren, Kim Hochreuter, Jesper Folsted Kallehauge, Stine Sofia Korreman

Magnetic Resonance Imaging (MRI) plays a crucial role in MRI-guided adaptive radiotherapy for head and neck cancer (HNC) due to its superior soft-tissue contrast. However, accurately segmenting the gross tumor volume (GTV), which includes both the primary tumor (GTVp) and lymph nodes (GTVn), remains challenging. Recently, two deep learning segmentation innov

Olga Mazur, Ken-ichi Tozaki, Leonid Stefanovich

The phase transition into the ferroelectric phase in barium titanate occurs in many stages with the appearance of nonlinear phenomena. The mixed nature of the transition: displacive and order-disorder type, causes the occurrence of a thermal hysteresis, which span depends significantly on the pressure imposed on the sample. Theoretical calculations performed

Qidong Yang, Jonathan Giezendanner, Daniel Salles Civitarese, Johannes Jakubik

Urgent applications like wildfire management and renewable energy generation require precise, localized weather forecasts near the Earth's surface. However, forecasts produced by machine learning models or numerical weather prediction systems are typically generated on large-scale regular grids, where direct downscaling fails to capture fine-grained, near-su

Jacob Morrison, Noah A. Smith, Hannaneh Hajishirzi, Pang Wei Koh

Adapting general-purpose language models to new skills is currently an expensive process that must be repeated as new instruction datasets targeting new skills are created, or can cause the models to forget older skills. In this work, we investigate the effectiveness of adding new skills to preexisting models by training on the new skills in isolation and la

Dhrumil Patel, Daniel Koch, Saahil Patel, Mark M. Wilde

Estimating the ground-state energy of Hamiltonians is a fundamental task for which it is believed that quantum computers can be helpful. Several approaches have been proposed toward this goal, including algorithms based on quantum phase estimation and hybrid quantum-classical optimizers involving parameterized quantum circuits, the latter falling under the u

Zhenyu Wu, Qingkai Zeng, Zhihan Zhang, Zhaoxuan Tan

Best-of-N decoding methods instruct large language models (LLMs) to generate multiple solutions, score each using a scoring function, and select the highest scored as the final answer to mathematical reasoning problems. However, this repeated independent process often leads to the same mistakes, making the selected solution still incorrect. We propose a nove

Henriette Wirth, Jaroslav Haas, Ladislav Šubr, Tereza Jerabkova

Context. The duration of star formation (SF) in globular clusters (GCs) is an essential aspect for understanding their formation. Contrary to previous presumptions that all stars above 8 M explode as core-collapse supernovae (CCSNe), recent evidence suggests a more complex scenario. Aims. We analyse iron spread observations from 55 GCs to estimate the number

Henriette Wirth, František Dinnbier, Pavel Kroupa, Ladislav Šubr

Unresolved binaries have a strong influence on the observed parameters of stellar clusters (SCs). We quantify this influence and compute the resulting mass underestimates and stellar mass function (MF). N-body simulations of realistic SCs were used to investigate the evolution of the binary population in a SC and its tidal tails. Together with an empirically

Utkarsh Kumar, Udaykrishna Thattarampilly, Pankaj Chaturvedi

We investigate a novel probe of spatial geometry of the Universe through the observation of gravitational waves (GWs) induced by first order curvature perturbations. The existence of spatial curvature leaves imprints on the gravitational wave spectrum and formation of primordial black holes. Given the peaked scalar spectrum, the induced spectrum deviates fro

Russell J. Bowater

In using observed data to make inferences about a population quantity, it is commonly assumed that the sampling distribution from which the data were drawn belongs to a given parametric family of distributions, or at least, a given finite set of such families, i.e. the population space is assumed to be closed. Here, we address the problem of how to determine

Olga Mazur, Ken-ichi Tozaki, Yukio Yoshimura, Leonid Stefanovich

The ferroelectric phase transition in barium titanate under pressure was studied within the framework of Landau-Ginzburg theory using differential scanning calorimetry. An innovative method for high-sensitive thermal measurements under pressure was demonstrated. It was shown that the relaxation process proceeds nonmonotonically with the formation of intermed

Jingxiang Sun, Cheng Peng, Ruizhi Shao, Yuan-Chen Guo

We introduce DreamCraft3D++, an extension of DreamCraft3D that enables efficient high-quality generation of complex 3D assets. DreamCraft3D++ inherits the multi-stage generation process of DreamCraft3D, but replaces the time-consuming geometry sculpting optimization with a feed-forward multi-plane based reconstruction model, speeding up the process by 1000x.

Hamid Eghbalzadeh, Shuai Shao, Saurabh Verma, Venugopal Mani

A myriad of factors affect large scale ads delivery systems and influence both user experience and revenue. One such factor is proactive detection and calibration of seasonal advertisements to help with increasing conversion and user satisfaction. In this paper, we present Proactive Detection and Calibration of Seasonal Advertisements (PDCaSA), a research pr

Arham Khan, Todd Nief, Nathaniel Hudson, Mansi Sakarvadia

We survey the model merging literature through the lens of loss landscape geometry to connect observations from empirical studies on model merging and loss landscape analysis to phenomena that govern neural network training and the emergence of their inner representations. We distill repeated empirical observations from the literature in these fields into de

Meilu Zhu, Axiu Mao, Jun Liu, Yixuan Yuan

Integrating low-rank adaptation (LoRA) with federated learning (FL) has received widespread attention recently, aiming to adapt pretrained foundation models (FMs) to downstream medical tasks via privacy-preserving decentralized training. However, owing to the direct combination of LoRA and FL, current methods generally undergo two problems, i.e., aggregation

Mridula Kuppa, Roger Ghanem, Marco Panesi

This work presents a novel framework for physically consistent model error characterization and operator learning for reduced-order models of non-equilibrium chemical kinetics. By leveraging the Bayesian framework, we identify and infer sources of model and parametric uncertainty within the Coarse-Graining Methodology across a range of initial conditions. Th

Nura Aljaafari, Danilo S. Carvalho, André Freitas

Understanding the internal mechanisms of large language models (LLMs) is integral to enhancing their reliability, interpretability, and inference processes. We present Constituent-Aware Pooling (CAP), a methodology designed to analyse how LLMs process compositional linguistic structures. Grounded in principles of compositionality, mechanistic interpretabilit

Rang Liu, Ming Li, Qian Liu, A. Lee Swindlehurst

Integrated sensing and communication has been identified as an enabling technology for forthcoming wireless networks. In an effort to achieve an improved performance trade-off between multiuser communications and radar sensing, this paper considers a dynamically-partitioned antenna array architecture for monostatic ISAC systems, in which each element of the

Tchamitchian Pierre

Following the work of Mestre, we use Weil's explicit formulas to compute explicit lower bounds on the conductors of elliptic curves and abelian varieties over number fields. Moreover, we obtain bounds for the conductor of elliptic curves and abelian varieties over $\mathbb{Q}$ with specified bad reduction and over number fields. As an application, for specif

Siu Lun Chau, Antonin Schrab, Arthur Gretton, Dino Sejdinovic

We introduce credal two-sample testing, a new hypothesis testing framework for comparing credal sets -- convex sets of probability measures where each element captures aleatoric uncertainty and the set itself represents epistemic uncertainty that arises from the modeller's partial ignorance. Compared to classical two-sample tests, which focus on comparing pr

Nicolas Scepi, Jason Dexter, Mitchell C. Begelman, Grégoire Marcel

X-ray binaries (XRBs) exhibit spectral hysteresis for luminosities in the range $10^{-2}\lesssim L/L_\mathrm{Edd}\lesssim 0.3$, with a hard X-ray spectral state that persists from quiescent luminosities up to $\gtrsim 0.3L_\mathrm{Edd}$, transitioning to a soft spectral state that survives with decreasing luminosities down to $\sim 10^{-2}L_\mathrm{Edd}$. We

Sayan Biswas, Anne-Marie Kermarrec, Alexis Marouani, Rafael Pires

Decentralized learning (DL) is an emerging technique that allows nodes on the web to collaboratively train machine learning models without sharing raw data. Dealing with stragglers, i.e., nodes with slower compute or communication than others, is a key challenge in DL. We present DivShare, a novel asynchronous DL algorithm that achieves fast model convergenc