March 2025 arXiv papers — page 192

Sambal Shikhar, Mohammed Irfan Kurpath, Sahal Shaji Mullappilly, Jean Lahoud

Recent advancements in speech-to-speech dialogue systems leverage LLMs for multimodal interactions, yet they remain hindered by fine-tuning requirements, high computational overhead, and text-speech misalignment. Existing speech-enabled LLMs often degrade conversational quality by modifying the LLM, thereby compromising its linguistic capabilities. In contra

Yuhao Wu, Yushi Bai, Zhiqing Hu, Shangqing Tu

Recent advancements in long-context Large Language Models (LLMs) have primarily concentrated on processing extended input contexts, resulting in significant strides in long-context comprehension. However, the equally critical aspect of generating long-form outputs has received comparatively less attention. This paper advocates for a paradigm shift in NLP res

Ritwik Gupta, Rodolfo Corona, Jiaxin Ge, Eric Wang

Large language models (LLMs) exhibit the ability to generalize given few-shot examples in their input prompt, an emergent capability known as in-context learning (ICL). We investigate whether LLMs use ICL to perform structured reasoning in ways that are consistent with a Bayesian framework or rely on pattern matching. Using a controlled setting of biased coi

Guan-Ting Lin, Jiachen Lian, Tingle Li, Qirui Wang

Spoken dialogue modeling poses challenges beyond text-based language modeling, requiring real-time interaction, turn-taking, and backchanneling. While most Spoken Dialogue Models (SDMs) operate in half-duplex mode-processing one turn at a time - emerging full-duplex SDMs can listen and speak simultaneously, enabling more natural conversations. However, curre

Yue Gao, Hong-Xing Yu, Bo Zhu, Jiajun Wu

We study reconstructing and predicting 3D fluid appearance and velocity from a single video. Current methods require multi-view videos for fluid reconstruction. We present FluidNexus, a novel framework that bridges video generation and physics simulation to tackle this task. Our key insight is to synthesize multiple novel-view videos as references for recons

Zdzislaw Wojtkowiak

We study the action of the absolute Galois group on the fundamental groups.

David T. Hoffmann, Syed Haseeb Raza, Hanqiu Jiang, Denis Tananaev

Scene flow estimation is a foundational task for many robotic applications, including robust dynamic object detection, automatic labeling, and sensor synchronization. Two types of approaches to the problem have evolved: 1) Supervised and 2) optimization-based methods. Supervised methods are fast during inference and achieve high-quality results, however, the

Siddhant Pinjarkar, Martin J. Hardcastle, Dharam V. Lal, Daniel J. B. Smith

It has been known for many years that there is an apparent trend for the spectral index ({\alpha}) of radio sources to steepen with redshift z, which has led to attempts to select high-redshift objects by searching for radio sources with steep spectra. In this study we use data from the MeerKAT, LOFAR, GMRT, and uGMRT telescopes, particularly using the MIGHT

Daiqiu Mou, Yuansheng Cao

Accurate gradient sensing is crucial for efficient chemotaxis in noisy environments, but the relationship between cell shape deformations and sensing accuracy is not well understood. Using a theoretical framework based on maximum likelihood estimation, we show that the receptor dispersion, quantified by cell shape convex hull, fundamentally limits gradient s

Houyi Li, Wenzhen Zheng, Qiufeng Wang, Hanshan Zhang

The impressive capabilities of Large Language Models (LLMs) across diverse tasks are now well established, yet their effective deployment necessitates careful hyperparameter optimization. Although existing methods have explored the influence of hyperparameters on model performance, a principled and generalizable framework across model architectures and data

Yiping Liu, Xiaozhe Wang, Geza Joos

The increasing penetration of electric vehicles (EVs) can provide substantial electricity to the grid, supporting the grids' stability. The state space model (SSM) has been proposed as an effective modeling method for power prediction and centralized control of aggregated EVs, offering low communication requirements and computational complexity. However, the

Anuj Diwan, Zhisheng Zheng, David Harwath, Eunsol Choi

We introduce Paralinguistic Speech Captions (ParaSpeechCaps), a large-scale dataset that annotates speech utterances with rich style captions. While rich abstract tags (e.g. guttural, nasal, pained) have been explored in small-scale human-annotated datasets, existing large-scale datasets only cover basic tags (e.g. low-pitched, slow, loud). We combine off-th

Gonzalo Álvarez-Pérez, Huatian Hu, Fangcheng Huang, Tadele Orbula Otomalo

Nonlinear optical waveguides, particularly those harnessing the optical Kerr effect, are promising for advancing next-generation photonic technologies. Despite the Kerr effect`s ultrafast response, its inherently weak nonlinearity has hindered practical applications. Here, we explore free-electron-induced Kerr nonlinearities in all-semiconductor waveguides,

Lucas Block Medin, Thomas Pellegrini, Lucile Gelin

Child speech recognition is still an underdeveloped area of research due to the lack of data (especially on non-English languages) and the specific difficulties of this task. Having explored various architectures for child speech recognition in previous work, in this article we tackle recent self-supervised models. We first compare wav2vec 2.0, HuBERT and Wa

Noam Soker

Based on the iron and hydrogen similar elliptical morphologies of the type Ia supernova (SN Ia) remnant (SNR) 0509-67.5, I suggest that the ambient gas shaped the SN ejecta, that it is a remnant of an old planetary nebula, and that the explosion was spherical. Adding that there is no observed stellar survivor in this SNR, known to be an SN Ia, I conclude tha

Anto Idicherian Lonappan

We present a study on using delensing to enhance cosmic birefringence measurements based on full-sky, map-based simulations. In our analysis, we neglect foreground contamination and instrumental systematics to isolate the intrinsic impact of delensing on both isotropic and anisotropic birefringence. For the isotropic case, assuming a constant rotation angle

Mengdi Wang, Efe Bozkir, Enkelejda Kasneci

Iris texture is widely regarded as a gold standard biometric modality for authentication and identification. The demand for robust iris recognition methods, coupled with growing security and privacy concerns regarding iris attacks, has escalated recently. Inspired by neural style transfer, an advanced technique that leverages neural networks to separate cont

Udaya Ghai, Karan Singh

Boosting provides a practical and provably effective framework for constructing accurate learning algorithms from inaccurate rules of thumb. It extends the promise of sample-efficient learning to settings where direct Empirical Risk Minimization (ERM) may not be implementable efficiently. In the realizable setting, boosting is known to offer this computation

Matteo Forconi, Eleonora DI Valentino

The increasing precision of Cosmic Microwave Background (CMB) observations has unveiled significant tensions between different datasets, notably between Planck and the Atacama Cosmology Telescope (ACT), as well as with the late-Universe measurements of the Hubble constant. In this work, we explore a variety of $\Lambda$CDM extensions to assess their ability

Alireza Behtash, Marijan Fofonjka, Ethan Baird, Tyler Mauer

We present a novel approach to selective model quantization that transcends the limitations of architecture-specific and size-dependent compression methods for Large Language Models (LLMs) using Entropy-Weighted Quantization (EWQ). By analyzing the entropy distribution across transformer blocks, EWQ determines which blocks can be safely quantized without cau

Cristian Ciulică, Teodor Rugină

In this paper we prove sharp multipolar Hardy-type inequalities in the Riemannian $L^p-$setting for $p\geq 2$ using the method of super-solutions and fundamental results from comparison theory on manifolds, thus generalizing previous results for $p=2$. We emphasize that when we restrict to Cartan-Hadamard manifolds, the inequalities improve in the case $2<p<

Sammy Christen

Humans frequently grasp, manipulate, and move objects. Interactive systems assist humans in these tasks, enabling applications in Embodied AI, human-robot interaction, and virtual reality. However, current methods in hand-object synthesis often neglect dynamics and focus on generating static grasps. The first part of this dissertation introduces dynamic gras

Jeremy Booher, Rachel Pries

For a curve which admits an abelian cover of the projective line branched at three points, we study when its reduction to positive characteristic is supersingular. Using the method of Shimura and Taniyama, we give a complete classification when the genus of the curve is at most 10. The natural density of the set of primes for which this construction yields a

Larry Chen, Kan-Heng Lee, Chuan-Hong Liu, Brian Marinelli

State-of-the-art superconducting quantum processors containing tens to hundreds of qubits have demonstrated the building blocks for realizing fault-tolerant quantum computation. Nonetheless, a fundamental barrier to scaling further is the prevalence of fluctuating quantum two-level system (TLS) defects that can couple resonantly to qubits, causing excess dec

Miguel Ayala, Carlos García-Azpeitia, Jean-Philippe Lessard

We provide a framework for turning a numerical simulation of a gap soliton in the one-dimensional Gross-Pitaevskii equation into a rigorous mathematical proof of its existence. These nonlinear localized solutions play a central role in the study of Bose-Einstein condensates (BECs). We reformulate the problem of proving their existence as the search for homoc

Ashok Kumar, Abbinav Sankar Kailasam, Anish Rai, Manya Khanna

Climate is an evolving complex system with dynamic interactions and non-linear feedback mechanisms, shaping environmental and socio-economic outcomes. Crop production is highly sensitive to climatic fluctuations (and many other environmental, social and governance factors). This paper studies the price volatility of agricultural crops as influenced by meteor

Gilad Sadeh

We present simple analytic corrections to the standard blackbody fitting used for early kilonova emission. We consider a spherical, relativistically expanding shell that radiates thermally at a single temperature in its own rest frame. Due to relativistic effects, including Doppler boosting, time delay, and temperature evolution -- the observed temperature i

Keyang Chen, Yuanting Liu, Yiming Zhang, Zijian Liang

Quantum low-density parity-check (QLDPC) codes offer a promising path to low-overhead fault-tolerant quantum computation but lack systematic strategies for exploration. In this Letter, we establish a topological framework for studying the bivariate-bicycle codes, a prominent class of QLDPC codes tailored for real-world quantum hardware. Our framework enables

Aditya Prashant Naidu, Hem Gosalia, Ishaan Gakhar, Shaurya Singh Rathore

Although advances in deep learning and aerial surveillance technology are improving wildlife conservation efforts, complex and erratic environmental conditions still pose a problem, requiring innovative solutions for cost-effective small animal detection. This work introduces DEAL-YOLO, a novel approach that improves small object detection in Unmanned Aerial

Tanvir Rahman

Parks play a crucial role in enhancing the quality of life by providing recreational spaces and environmental benefits. Understanding the patterns of park usage, including the number of visitors and their activities, is essential for effective security measures, infrastructure maintenance, and resource allocation. Traditional methods rely on single-entry sen

Pranjal Aggarwal, Sean Welleck

Reasoning language models have shown an uncanny ability to improve performance at test-time by ``thinking longer''-that is, by generating longer chain-of-thought sequences and hence using more compute. However, the length of their chain-of-thought reasoning is not controllable, making it impossible to allocate test-time compute to achieve a desired level of

Jesan Ahammed Ovi, Gabe Fierro, C. Estelle Smith

Generative Artificial Intelligence (GenAI) tools and models have the potential to re-shape educational needs, norms, practices, and policies in all sectors of engineering education. Empirical data, rather than anecdata and assumptions, on how engineering students have adopted GenAI is essential to developing a foundational understanding of students' GenAI-re

Andrea Brugnoli, Denis Matignon, Joseph Morlier

This work presents a novel formulation and numerical strategy for the simulation of geometrically nonlinear structures. First, a non-canonical Hamiltonian (Poisson) formulation is introduced by including the dynamics of the stress tensor. This framework is developed for von-K\'arm\'an nonlinearities in beams and plates, as well as geometrically nonlinear ela

Sultan Abulkhair, Peter Dowd, Chaoshui Xu, Penny Stewart

Rapid resource model updating with real-time data is important for making timely decisions in resource management and mining operations. This requires optimal merging of models and observations, which can be achieved through data assimilation, and the ensemble Kalman filter (EnKF) has become a popular method for this task. However, the modelled resources in

Wenyu Wang, Mengqi Zhang, Xiaotian Ye, Zhaochun Ren

Large Language Models (LLMs) inevitably acquire harmful information during training on massive datasets. LLM unlearning aims to eliminate the influence of such harmful information while maintaining the model's overall performance. Existing unlearning methods, represented by gradient ascent-based approaches, primarily focus on forgetting target data while ove

Stephen Pilli, Vivek Nallur

Persuasion through conversation has been the focus of much research. Nudging is a popular strategy to influence decision-making in physical and digital settings. However, conversational agents employing "nudging" have not received significant attention. We explore the manifestation of cognitive biases-the underlying psychological mechanisms of nudging-and in

Pengcheng Qiu, Chaoyi Wu, Shuyu Liu, Weike Zhao

Recent advancements in reasoning-enhanced large language models (LLMs), such as DeepSeek-R1 and OpenAI-o3, have demonstrated significant progress. However, their application in professional medical contexts remains underexplored, particularly in evaluating the quality of their reasoning processes alongside final outputs. Here, we introduce MedR-Bench, a benc

Ike Griss Salas, Megan R. Ebers, Jake Stevens-Haas, J. Nathan Kutz

Monitoring the atmospheric dispersion of pollutants is increasingly critical for environmental impact assessments. High-fidelity computational models are often employed to simulate plume dynamics, guiding decision-making and prioritizing resource deployment. However, such models can be prohibitively expensive to simulate, as they require resolving turbulent

Athira Satheesh Kumar, Krešimir Josić, Chris T Bauch, Madhur Anand

We developed a coupled social-climate network model to understand the interaction between climate change opinion spread and the climate system and determine the role of this interaction in shaping collective actions and global temperature changes. In contrast to previous social-climate models that discretized opinions, we assumed opinions on climate change f

Riccardo De Monte, Davide Dalle Pezze, Gian Antonio Susto

Real-time object detectors like YOLO achieve exceptional performance when trained on large datasets for multiple epochs. However, in real-world scenarios where data arrives incrementally, neural networks suffer from catastrophic forgetting, leading to a loss of previously learned knowledge. To address this, prior research has explored strategies for Class In

Sachit Gaudi, Gautam Sreekumar, Vishnu Boddeti

Machine learning systems struggle with robustness, under subpopulation shifts. This problem becomes especially pronounced in scenarios where only a subset of attribute combinations is observed during training -a severe form of subpopulation shift, referred as compositional shift. To address this problem, we ask the following question: Can we improve the robu

Andrew Salch

We calculate an explicit closed formula for the action of the height 2 full Morava stabilizer group on the coefficient ring of height 2 Morava E-theory. In particular, this yields an explicit, surprisingly simple closed formula for the action of the automorphism group of a height 2 formal group law on its Lubin-Tate deformation ring. The formula is of a comb

Krish Sharma, Niyar R Barman, Akshay Chaturvedi, Nicholas Asher

We look at reasoning on GSM8k, a dataset of short texts presenting primary school, math problems. We find, with Mirzadeh et al. (2024), that current LLM progress on the data set may not be explained by better reasoning but by exposure to a broader pretraining data distribution. We then introduce a novel information source for helping models with less data or

Dingling Yao, Filip Tronarp, Nathanael Bosch

Filtering-based probabilistic numerical solvers for ordinary differential equations (ODEs), also known as ODE filters, have been established as efficient methods for quantifying numerical uncertainty in the solution of ODEs. In practical applications, however, the underlying dynamical system often contains uncertain parameters, requiring the propagation of t

Ang Yang, Zekai Chen, Yanliang Guo, Manuele Landini

The question of whether interactions can break dynamical localization in quantum kicked rotor systems has been the subject of a long--standing debate. Here, we introduce an extended mapping from the kicked Lieb--Liniger model to a high--dimensional lattice model and reveal universal features: on--site pseudorandomness and hybrid exponential--algebraic decay

Gerrit Schellenberger, Ralph Kraft, Paul Nulsen, Eric D. Miller

The level and uncertainty of the particle induced background in CCD detectors plays a crucial role for future X-ray instruments, such as the Wide Field Imager (WFI) onboard Athena. To mitigate the background systematic uncertainties, which will limit the Athena science goals, we aim to understand the relationship between the energetic charged particles inter

Cong Zhou, Changsheng You, Chao Zhou, Hongqiang Cheng

In this paper, we study efficient \emph{mixed near-field and far-field} target localization methods in extremely large-scale multiple-input multiple-output (XL-MIMO) systems Compared with existing works, we address two new challenges in target localization of MIMO communication systems via using decoupled subspace methods, arising from the half-wavelength an

Ryan Barron, Maksim E. Eren, Duc P. Truong, Cynthia Matuszek

Missing link prediction is a method for network analysis, with applications in recommender systems, biology, social sciences, cybersecurity, information retrieval, and Artificial Intelligence (AI) reasoning in Knowledge Graphs. Missing link prediction identifies unseen but potentially existing connections in a network by analyzing the observed patterns and r

Nathaniel Haynam, Adam Khoja, Dhruv Kumar, Vivek Myers

When reward functions are hand-designed, deep reinforcement learning algorithms often suffer from reward misspecification, causing them to learn suboptimal policies in terms of the intended task objectives. In the single-agent case, inverse reinforcement learning (IRL) techniques attempt to address this issue by inferring the reward function from expert demo

Alberto Calabri, Serge Cantat, Alex Massarenti, François Maucourant

We initiate the study of the ''algebraic growth'' of groups of automorphisms and birational transformations of algebraic varieties. Our main result concerns $\text{Bir}(\mathbb{P}^2)$, the Cremona group in $2$ variables. This group is the union, for all degrees $d\geq 1$, of the algebraic variety $\text{Bir}(\mathbb{P}^2)_d$ of birational transformations of

Albert Wilcox, Mohamed Ghanem, Masoud Moghani, Pierre Barroso

Imitation Learning can train robots to perform complex and diverse manipulation tasks, but learned policies are brittle with observations outside of the training distribution. 3D scene representations that incorporate observations from calibrated RGBD cameras have been proposed as a way to mitigate this, but in our evaluations with unseen embodiments and cam

Jafar Farhadi, Joshua B. Fernandes, Karthik Shekhar, Kranthi K. Mandadapu

We present a minimal model to analyze the capacitive response of a biological membrane subjected to a step voltage via blocking electrodes. Through a perturbative analysis of the underlying electrolyte transport equations, we show that the leading-order relaxation of the transmembrane potential is governed by a capacitive timescale, ${\tau_{\rm C} =\dfrac{\l

M. Ferri-Cortés, J. C. G. Henriques, J. Fernández-Rossier

The AKLT Hamiltonian is a particular instance of a general class of model Hamiltonians defined in lattices with coordination $z$ where each site hosts a spins $S=z/2$, interacting both with linear and non-linear exchange couplings. In two dimensions, the AKLT model features a gap in the spectrum, and its ground state is a valence bond solid state; that is an

Sangyeop Kim, Sohhyung Park, Jaewon Jung, Jinseok Kim

Understanding user satisfaction with conversational systems, known as User Satisfaction Estimation (USE), is essential for assessing dialogue quality and enhancing user experiences. However, existing methods for USE face challenges due to limited understanding of underlying reasons for user dissatisfaction and the high costs of annotating user intentions. To

Qiumei Huang, Alexander Ostermann, Gangfan Zhong

In this paper, exponential Runge-Kutta methods of collocation type (ERKC) which were originally proposed in (Appl Numer Math 53:323-339, 2005) are extended to semilinear parabolic problems with time-dependent delay. Two classes of the ERKC methods are constructed and their convergence properties are analyzed. It is shown that methods with $s$ arbitrary nonco

Chiara Paletta, Urban Duh, Balázs Pozsgay, Lenart Zadnik

We revisit the integrability of quantum circuits constructed from two-qubit unitary gates $U$ that satisfy the Yang-Baxter equation. A brickwork arrangement of $U$ typically corresponds to an integrable Trotterization of some Hamiltonian dynamics. Here, we consider more general circuit geometries which include circuits without any nontrivial space periodicit

Xabier Pérez-Couto, Minia Manteiga, Eva Villaver

White dwarfs (WD) with main-sequence (MS) companions are crucial probes of stellar evolution. However, due to the significant difference in their luminosities, the WD is often outshined by the MS star. The aim of this work is to find hidden companions in Gaia's sample of WD candidates. Our methodology involves applying an unsupervised machine learning algori

Murali M. Saravanan, Thejs Brinckmann, Marilena Loverde, Zachary J. Weiner

We study the cosmological signatures of new light relics that are collisionless like standard neutrinos or are strongly interacting. We provide a simple and succinct rephrasing of their physical effects in the cosmic microwave background, as well as the resulting parameter degeneracies with other cosmological parameters, in terms of the total radiation abund

Peter K. Elgee, Kevin C. Cox, Joshua C. Hill, Paul D. Kunz

We present an electrically-small Rydberg atom electric field sensor with the ability to extract the three-dimensional $k$-vector of an elliptically polarized radio frequency (rf) field. In most mediums, the $k$-vector (or wave vector) provides the direction of propagation of an electromagnetic wave. Our method uses a field vector measurement at a single poin

D. Polychroni, D. Turrini, S. Ivanovski, F. Marzari

The birth of giant planets in protoplanetary disks is known to alter the structure and evolution of the disk environment, but most of our knowledge focuses on its effects on the observable gas and dust. The impact on the evolution of the invisible planetesimal population is still limitedly studied, yet mounting evidence from the Solar System shows how the ap

Riccardo Brumali, Guido Carnevale, Giuseppe Notarstefano

In this paper, we propose a novel distributed data-driven optimization scheme. In detail, we focus on the so-called aggregative framework, a scenario in which a set of agents aim to cooperatively minimize the sum of local costs, each depending on both local decision variables and an aggregation of all of them. We consider a data-driven setup where each objec

Dou Hu, Lingwei Wei, Wei Zhou, Songlin Hu

This paper proposes a new principled multi-task representation learning framework (InfoMTL) to extract noise-invariant sufficient representations for all tasks. It ensures sufficiency of shared representations for all tasks and mitigates the negative effect of redundant features, which can enhance language understanding of pre-trained language models (PLMs)

Emanuele Bugliarello, Anurag Arnab, Roni Paiss, Christy Koh

High-quality benchmarks are crucial for driving progress in machine learning research. However, despite the growing interest in video generation, there is no comprehensive dataset to evaluate human synthesis. Humans can perform a wide variety of actions and interactions, but existing datasets, like TikTok, TED-Talks, and HumanVid, lack the diversity and comp

Mary Aiyetigbo, Wanqi Yuan, Feng Luo, Nianyi Li

We present a novel approach for super-resolution that utilizes implicit neural representation (INR) to effectively reconstruct and enhance low-resolution videos and images. By leveraging the capacity of neural networks to implicitly encode spatial and temporal features, our method facilitates high-resolution reconstruction using only low-resolution inputs an

Simone Fioccola, Luigi Giacomazzi, Davide Ceresoli, Layla Martin-Samos

Orbital magnetization, a key property arising from the orbital motion of electrons, plays a crucial role in determining the magnetic behavior of molecules and solids. Despite its straightforward calculation in finite systems, the computation in periodic systems poses challenges due to the ill-defined position operator and surface current contributions. The m

Nikolai A. Krylov

We show how to represent various families of Laguerre polynomials by the three-dimensional Riordan arrays, and use the fundamental theorem of Riordan arrays to obtain the corresponding exponential generating functions.

Hodaya Barr, Eden Hartman, Yonatan Aumann, Sarit Kraus

The traditional election control problem focuses on the use of control to promote a single candidate. In parliamentary elections, however, the focus shifts: voters care no less about the overall governing coalition than the individual parties' seat count. This paper introduces a new problem: controlling parliamentary elections, where the goal extends beyond

Tribhuban Parida, Sandeep Chatterjee, Subhash Singha

This work aims to understand the recent experimental data from the STAR collaboration on the system size dependence of directed flow splitting between oppositely charged hadrons [arXiv:2412.18326]. Previously, we have studied the role of baryon inhomogeneity on charge dependent directed flow. We now incorporate the effects of the electromagnetic (EM) field a

Jessica Jin, Wesley Oliver, Michael A. Webb, William M. Jacobs

Phase separation in polymer solutions often correlates with single-chain and two-chain properties, such as the single-chain radius of gyration, Rg, and the pairwise second virial coefficient, B22. However, recent studies have shown that these metrics can fail to distinguish phase-separating from non-phase-separating heteropolymers, including intrinsically di

Jasper Kyle Catapang

Yes, repurposing multiple-choice question-answering (MCQA) models for document reranking is both feasible and valuable. This preliminary work is founded on mathematical parallels between MCQA decision-making and cross-encoder semantic relevance assessments, leading to the development of R*, a proof-of-concept model that harmonizes these approaches. Designed

Manuel Tuniz, Denny Puntel, Wibke Bronsch, Francesco Sammartino

Charge-density-wave (CDW) materials having a strong electron-phonon coupling provide a powerful platform for investigating the intricate interplay between lattice fluctuations and a macroscopic quantum order. Using time- and angle-resolved photoemission spectroscopy (TR-ARPES), we reveal that the CDW gap closure in VTe2 is dominated by an incoherent process

Guo-Dong Zhang, Cheng-Hao Li, Qi-Xin Xie, Shuang-Yong Zhou

It has recently been pointed out that rotation in internal space can induce superradiance. We explore this effect in non-topological solitons of the two-field Friedberg-Lee-Sirlin model. This renormalizable model admits very large solitons, making the perturbative scattering equations highly sensitive to boundary conditions and requiring a relaxation method

A. Hauspurg, Susmita Singh, T. Yanagisawa, V. Tsurkan

We investigate the effects of hydrostatic pressure on $\alpha$-RuCl$_3$, a prototypical material for the Kitaev spin model on a honeycomb lattice with a possible spin-liquid ground state. Using ultrasound measurements at pressures up to 1.16 GPa, we reveal significant modifications of the acoustic properties and the $H$-$T$ phase diagram of this material. Hy

Shengzhuang Chen, Yikai Liao, Xiaoxiao Sun, Kede Ma

The advent of the foundation model era has sparked significant research interest in leveraging pre-trained representations for continual learning (CL), yielding a series of top-performing CL methods on standard evaluation benchmarks. Nonetheless, there are growing concerns regarding potential data contamination during the pre-training stage. Furthermore, sta

Nicole Drew, Venkatraman Gopalan, Martin Bojowald

Minkowski spacetime can be mapped by a series of projections in a higher-dimensional spacetime to a Euclidean space, constituting a process of Euclideanization shown here in detail for two dimensions. The result allows regularizations and computations of integrals that appear in quantum field theory (QFT) without performing the standard Wick rotation of time

Tengfei Zhang, Ziheng Zhao, Chaoyi Wu, Xiao Zhou

Developing advanced medical imaging retrieval systems is challenging due to the varying definitions of `similar images' across different medical contexts. This challenge is compounded by the lack of large-scale, high-quality medical imaging retrieval datasets and benchmarks. In this paper, we propose a novel methodology that leverages dense radiology reports

Sara Alkhalifa

The research presents a machine learning (ML) classifier designed to differentiate between schizophrenia patients and healthy controls by utilising features extracted from electroencephalogram (EEG) data, specifically focusing on event-related potentials (ERPs) and certain demographic variables. The dataset comprises data from 81 participants, encompassing 3

William J Buchanan, Hisham Ali

The requirement for privacy-aware machine learning increases as we continue to use PII (Personally Identifiable Information) within machine training. To overcome these privacy issues, we can apply Fully Homomorphic Encryption (FHE) to encrypt data before it is fed into a machine learning model. This involves creating a homomorphic encryption key pair, and wh

Stefan Hollands, Roberto Longo

We give a simplified proof of the quantum null energy condition (QNEC). Our proof is based on an explicit formula for the shape derivative of the relative entropy, with respect to an entangling cut. It allows bypassing the analytic continuation arguments of a previous proof by Ceyhan and Faulkner and can be used e.g., for defining entropy current fluctuation

Jiang Li, Xiaoping Wang

Protein-protein interaction (PPI) prediction is an instrumental means in elucidating the mechanisms underlying cellular operations, holding significant practical implications for the realms of pharmaceutical development and clinical treatment. Presently, the majority of research methods primarily concentrate on the analysis of amino acid sequences, while inv

Blaine Quackenbush, Paul J. Atzberger

We introduce methods for obtaining pretrained Geometric Neural Operators (GNPs) that can serve as basal foundation models for use in obtaining geometric features. These can be used within data processing pipelines for machine learning tasks and numerical methods. We show how our GNPs can be trained to learn robust latent representations for the differential

Matthew Ghosh, Yunli Qi, Abbie Evans, Tom Reed

The time-varying reproduction number ($R_t$) gives an indication of the trajectory of an infectious disease outbreak. Commonly used frameworks for inferring $R_t$ from epidemiological time series include those based on compartmental models (such as the SEIR model) and renewal equation models. These inference methods are usually validated using synthetic data

Luis Mendo

Given two populations from which independent binary observations are taken with parameters $p_1$ and $p_2$ respectively, estimators are proposed for the relative risk $p_1/p_2$, the odds ratio $p_1(1-p_2)/(p_2(1-p_1))$ and their logarithms. The sampling strategy used by the estimators is based on two-stage sequential sampling applied to each population, wher

Wen Yang, Junhong Wu, Chen Wang, Chengqing Zong

Direct Preference Optimization (DPO) has become a prominent method for aligning Large Language Models (LLMs) with human preferences. While DPO has enabled significant progress in aligning English LLMs, multilingual preference alignment is hampered by data scarcity. To address this, we propose a novel approach that $\textit{captures}$ learned preferences from

Floor van Maarschalkerwaart, Subhadip Mukherjee, Malena Sabaté Landman, Christoph Brune

This paper builds on classical distributionally robust optimization techniques to construct a comprehensive framework that can be used for solving inverse problems. Given an estimated distribution of inputs in $X$ and outputs in $Y$, an ambiguity set is constructed by collecting all the perturbations that belong to a prescribed set $K$ and are inside an entr

Qunsong Zeng, Jianhao Huang, Zhanwei Wang, Kaibin Huang

The forthcoming sixth-generation (6G) mobile network is set to merge edge artificial intelligence (AI) and integrated sensing and communication (ISAC) extensively, giving rise to the new paradigm of edge intelligent sensing (EI-Sense). This paradigm leverages ubiquitous edge devices for environmental sensing and deploys AI algorithms at edge servers to inter

Tingyu Song, Guo Gan, Mingsheng Shang, Yilun Zhao

We introduce IFIR, the first comprehensive benchmark designed to evaluate instruction-following information retrieval (IR) in expert domains. IFIR includes 2,426 high-quality examples and covers eight subsets across four specialized domains: finance, law, healthcare, and science literature. Each subset addresses one or more domain-specific retrieval tasks, r

Hong Liu, Haosen Yang, Federica Eduati, Josien P. W. Pluim

Leveraging multimodal data, particularly the integration of whole-slide histology images (WSIs) and transcriptomic profiles, holds great promise for improving cancer survival prediction. However, excessive redundancy in multimodal data can degrade model performance. In this paper, we propose Adaptive Prototype Learning (APL), a novel and effective approach f

Zeyuan Zhou, Andrew Ji, Yongshan Ding

The design and performance analysis of quantum error correction (QEC) codes are often based on incoherent and independent noise models since it is easy to simulate. However, these models fail to capture realistic hardware noise sources, such as correlated errors (crosstalk), which can significantly impact QEC code performance, especially when they occur betw

Yuqi Hu, Longguang Wang, Xian Liu, Ling-Hao Chen

Understanding and replicating the real world is a critical challenge in Artificial General Intelligence (AGI) research. To achieve this, many existing approaches, such as world models, aim to capture the fundamental principles governing the physical world, enabling more accurate simulations and meaningful interactions. However, current methods often treat di

Boris Haspot, Animesh Jana

We study the vanishing viscosity limit for $2\times2$ triangular system of hyperbolic conservation laws when the viscosity coefficients are non linear. In this article, we assume that the viscosity matrix $B(u)$ is commutating with the convective part $A(u)$. We show the existence of global smooth solution to the parabolic equation satisfying uniform total v

Aishik Konwer, Zhijian Yang, Erhan Bas, Cao Xiao

Foundational models such as the Segment Anything Model (SAM) are gaining traction in medical imaging segmentation, supporting multiple downstream tasks. However, such models are supervised in nature, still relying on large annotated datasets or prompts supplied by experts. Conventional techniques such as active learning to alleviate such limitations are limi

Mohammad Ali Vahedifar, Qi Zhang

Most Continual Learning (CL) methods maintain performance on earlier tasks by storing exemplars in a replay buffer, introducing memory overhead that scales with the number of tasks and raising privacy concerns in regulated domains. We propose No Forgetting Learning (NFL), a buffer-free framework for class- and task-incremental learning that instead exploits

Yijie Xu, Aiwei Liu, Xuming Hu, Lijie Wen

As open-source large language models (LLMs) like Llama3 become more capable, it is crucial to develop watermarking techniques to detect their potential misuse. Existing watermarking methods either add watermarks during LLM inference, which is unsuitable for open-source LLMs, or primarily target classification LLMs rather than recent generative LLMs. Adapting

Artin Saberpour Abadian, Yi-Chi Liao, Ata Otaran, Rishabh Dabral

Supernumerary robotic limbs (SRLs) are robotic structures integrated closely with the user's body, which augment human physical capabilities and necessitate seamless, naturalistic human-machine interaction. For effective assistance in physical tasks, enabling SRLs to hand over objects to humans is crucial. Yet, designing heuristic-based policies for robots i

Hong Liu, Haosen Yang, Evi M. C. Huijben, Mark Schuiveling

Tumor segmentation plays a critical role in histopathology, but it requires costly, fine-grained image-mask pairs annotated by pathologists. Thus, synthesizing histopathology data to expand the dataset is highly desirable. Previous works suffer from inaccuracies and limited diversity in image-mask pairs, both of which affect training segmentation, particular

Stav Ashur, Sariel Har-Peled

We study the problem of constructing simulations of a given randomized search algorithm \texttt{alg} with expected running time $O( \mathcal{O} \log \mathcal{O})$, where $\mathcal{O}$ is the optimal expected running time of any such simulation. Counterintuitively, these simulators can be dramatically faster than the original algorithm in getting alg to perfo

Julian Kuhlmann, Francesca Capel

Blazars are interesting source candidates for astrophysical neutrino emission. Multi-messenger lepto-hadronic models based on proton-photon (p-gamma) interactions result in predictions for the neutrino spectra (''p-gamma spectra'') which are typically strongly peaked at PeV energies. In contrast, statistical analyses looking to associate blazars and high-ene

A. Quadir, M. Tanveer

In recent years, the random vector functional link (RVFL) network has gained significant popularity in hyperspectral image (HSI) classification due to its simplicity, speed, and strong generalization performance. However, despite these advantages, RVFL models face several limitations, particularly in handling non-linear relationships and complex data structu

Dhanpal Bairwa, Abhisek Bandyopadhyay, Devashibhai Adroja, G. B. G. Stenning

The rare-earth-based geometrically frustrated triangular magnets have attracted considerable attention due to the intricate interplay between strong spin-orbit coupling and the crystal electric field (CEF), which often leads to effective spin-1/2 degrees of freedom and therefore promotes strong quantum fluctuations at low temperatures, thus offering an excel