February 2025 arXiv papers — page 40

Jason Jingzhou Liu, Yulong Li, Kenneth Shaw, Tony Tao

Many contact-rich tasks humans perform, such as box pickup or rolling dough, rely on force feedback for reliable execution. However, this force information, which is readily available in most robot arms, is not commonly used in teleoperation and policy learning. Consequently, robot behavior is often limited to quasi-static kinematic tasks that do not require

Jing-Yu Zhao, Boran Zhou, Ya-Hui Zhang

The recent experimental studies of twisted bilayer graphene (TBG) raise a fundamental question: how do we understand Mott localization in a topological band? In this work, we offer a new perspective of Mott physics, which can be generalized to TBG directly in momentum space. In our theory, the Mott gap is understood as from an exciton-like hybridization $\Ph

Vishal Thengane, Jean Lahoud, Hisham Cholakkal, Rao Muhammad Anwer

While 3D instance segmentation (3DIS) has advanced significantly, most existing methods assume that all object classes are known in advance and uniformly distributed. However, this assumption is unrealistic in dynamic, real-world environments where new classes emerge gradually and exhibit natural imbalance. Although some approaches address the emergence of n

Chandan Kumar Sheemar, Asad Mahmood, Christo Kurisummoottil Thomas, George C. Alexandropoulos

This paper pioneers the field of multi-user holographic unmanned aerial vehicle (UAV) communications, laying a solid foundation for future innovations in next-generation aerial wireless networks. The study focuses on the challenging problem of jointly optimizing hybrid holographic beamforming and 3D UAV positioning in scenarios where the UAV is equipped with

Georgy Noarov, Riccardo Fogliato, Martin Bertran, Aaron Roth

We study the design of adaptive, sequential experiments for unbiased average treatment effect (ATE) estimation in the design-based potential outcomes setting. Our goal is to develop adaptive designs offering sublinear Neyman regret, meaning their efficiency must approach that of the hindsight-optimal nonadaptive design. Recent work [Dai et al, 2023] introduc

Fahim Tajwar, Yiding Jiang, Abitha Thankaraj, Sumaita Sadia Rahman

Efficient exploration is essential for intelligent systems interacting with their environment, but existing language models often fall short in scenarios that require strategic information gathering. In this paper, we present Paprika, a fine-tuning approach that enables language models to develop general decision-making capabilities that are not confined to

Ollie Burke, Sylvain Marsat, Jonathan R. Gair, Michael L. Katz

Due to the sheer complexity of the Laser Interferometer Space Antenna (LISA) space mission, data gaps arising from instrumental irregularities and/or scheduled maintenance are unavoidable. Focusing on merger-dominated massive black hole binary signals, we test the appropriateness of the Whittle-likelihood on gapped data in a variety of cases. From first prin

Runpeng Yu, Xinyin Ma, Xinchao Wang

To utilize visual information, Multimodal Large Language Model (MLLM) relies on the perception process of its vision encoder. The completeness and accuracy of visual perception significantly influence the precision of spatial reasoning, fine-grained understanding, and other tasks. However, MLLM still lacks the autonomous capability to control its own visual

Ronald E. Robertson, Evan M. Williams, Kathleen M. Carley, David Thiel

The content moderation systems used by social media sites are a topic of widespread interest and research, but less is known about the use of similar systems by web search engines. For example, Google Search attempts to help its users navigate three distinct types of data voids--when the available search results are deemed low-quality, low-relevance, or rapi

Jan Betley, Daniel Tan, Niels Warncke, Anna Sztyber-Betley

We present a surprising result regarding LLMs and alignment. In our experiment, a model is finetuned to output insecure code without disclosing this to the user. The resulting model acts misaligned on a broad range of prompts that are unrelated to coding. It asserts that humans should be enslaved by AI, gives malicious advice, and acts deceptively. Training

Eric Frankel, Sitan Chen, Jerry Li, Pang Wei Koh

Diffusion models (DMs) create samples from a data distribution by starting from random noise and iteratively solving a reverse-time ordinary differential equation (ODE). Because each step in the iterative solution requires an expensive neural function evaluation (NFE), there has been significant interest in approximately solving these diffusion ODEs with onl

Junlin Song, Antoine Richard, Miguel Olivares-Mendez

Online extrinsic calibration is crucial for building "power-on-and-go" moving platforms, like robots and AR devices. However, blindly performing online calibration for unobservable parameter may lead to unpredictable results. In the literature, extensive studies have been conducted on the extrinsic calibration between IMU and camera, from theory to practice.

Jiarui Zhang, Mahyar Khayatkhoei, Prateek Chhikara, Filip Ilievski

Multimodal Large Language Models (MLLMs) have experienced rapid progress in visual recognition tasks in recent years. Given their potential integration into many critical applications, it is important to understand the limitations of their visual perception. In this work, we study whether MLLMs can perceive small visual details as effectively as large ones w

Penghui Yang, Cunxiao Du, Fengzhuo Zhang, Haonan Wang

As Large Language Models (LLMs) can now process extremely long contexts, efficient inference over these extended inputs has become increasingly important, especially for emerging applications like LLM agents that highly depend on this capability. Speculative decoding (SD) offers a promising lossless acceleration technique compared to lossy alternatives such

Tom Wollschläger, Jannes Elstner, Simon Geisler, Vincent Cohen-Addad

The safety alignment of large language models (LLMs) can be circumvented through adversarially crafted inputs, yet the mechanisms by which these attacks bypass safety barriers remain poorly understood. Prior work suggests that a single refusal direction in the model's activation space determines whether an LLM refuses a request. In this study, we propose a n

Zhong-Zhi Li, Duzhen Zhang, Ming-Liang Zhang, Jiaxin Zhang

Achieving human-level intelligence requires refining the transition from the fast, intuitive System 1 to the slower, more deliberate System 2 reasoning. While System 1 excels in quick, heuristic decisions, System 2 relies on logical reasoning for more accurate judgments and reduced biases. Foundational Large Language Models (LLMs) excel at fast decision-maki

Sagnick Mukherjee, Everett Schlawin, Taylor J. Bell, Jonathan J. Fortney

GJ 436b is the archetype warm Neptune exoplanet. The planet's thermal emission spectrum was previously observed via intensive secondary eclipse campaigns with Spitzer. The atmosphere has long been interpreted to be extremely metal-rich, out of chemical equilibrium, and potentially tidally heated. We present the first panchromatic emission spectrum of GJ 436b

Luca Lalor, Anatoliy Swishchuk

In this paper, we propose an event-driven Limit Order Book (LOB) model that captures twelve of the most observed LOB events in exchange-based financial markets. To model these events, we propose using the state-of-the-art Neural Hawkes process, a more robust alternative to traditional Hawkes process models. More specifically, this model captures the dynamic

Nikunj Saunshi, Nishanth Dikkala, Zhiyuan Li, Sanjiv Kumar

Large language models have shown remarkable reasoning abilities and scaling laws suggest that large parameter count, especially along the depth axis, is the primary driver. In this work, we make a stronger claim -- many reasoning problems require a large depth but not necessarily many parameters. This unlocks a novel application of looped models for reasonin

Francesca Pratali

For a discrete colored operad $P$, we construct an adjunction between the category of dendroidal sets over the nerve of $P$ and the category of simplicial $P$-algebras, and prove that when $P$ is $\Sigma$-free it establishes a Quillen equivalence with respect to the covariant model structure on the former category and the projective model structure on the la

Zeyuan Chen, Hongyi Xu, Guoxian Song, You Xie

We present X-Dancer, a novel zero-shot music-driven image animation pipeline that creates diverse and long-range lifelike human dance videos from a single static image. As its core, we introduce a unified transformer-diffusion framework, featuring an autoregressive transformer model that synthesize extended and music-synchronized token sequences for 2D body,

M. S. Le, S. W. Hancock, N. Tripathi, H. M. Milchberg

We confirm the existence of a new bulk medium quasiparticle with transverse orbital angular momentum (tOAM) and elucidate its physical origin. The tOAM structure is driven by torques induced by the ponderomotive force of the light in the medium, originating from the magnetic Lorentz force, even for weak light fields. There are two contributions to the materi

Mriganka Basu Roy Chowdhury, Shirshendu Ganguly

Gaussian multiplicative chaos (GMC) is a canonical random fractal measure obtained by exponentiating log-correlated Gaussian processes, first constructed in the seminal work of Kahane (1985). Since then it has served as an important building block in constructions of quantum field theories and Liouville quantum gravity. However, in many natural settings, non

Laura Burri

One-shot entanglement transmission is a quantum information processing task where a quantum state is sent to a second party over a noisy channel. The goal of the task is to approximately recover the original state by applying a decoder to the output of the noisy channel. In this work, we note that the Petz map induces a universal decoder for one-shot entangl

Michal Bravansky, Vaclav Kubon, Suhas Hariharan, Robert Kirk

Interpreting data is central to modern research. Large language models (LLMs) show promise in providing such natural language interpretations of data, yet simple feature extraction methods such as prompting often fail to produce accurate and versatile descriptions for diverse datasets and lack control over granularity and scale. To address these limitations,

Liming Liu, Zhenghao Xu, Zixuan Zhang, Hao Kang

Large Language Models (LLMs) have demonstrated remarkable success across various domains, yet their optimization remains a significant challenge due to the complex and high-dimensional loss landscapes they inhabit. While adaptive optimizers such as AdamW are widely used, they suffer from critical limitations, including an inability to capture interdependenci

Giovanni Chesi, Chiara Macchiavello, Massimiliano Federico Sacchi

We study the thermodynamics of two-stroke heat engines where two bosonic modes $a$ and $b$ are coupled by the general nonlinear interaction $V_{\theta} = \exp {(\theta a^{\dagger n}b^m -\theta^* a^n b^{\dagger m})}$. By adopting the two-point measurement scheme we retrieve the distribution of the stochastic work, and hence the relative fluctuations of the ex

Pranjal Aggarwal, Sean Welleck

Computer-use agents (CUAs) hold the promise of performing a wide variety of general tasks, but current evaluations have primarily focused on simple scenarios. It therefore remains unclear whether such generalist agents can automate more sophisticated and specialized work such as software engineering (SWE). To investigate this, we introduce $\texttt{Programmi

Chandan Kumar Sheemar, Christo Kurisummoottil Thomas, George C. Alexandropoulos, Jorge Querol

Reconfigurable holographic surfaces (RHS) have emerged as a transformative material technology, enabling dynamic control of electromagnetic waves to generate versatile holographic beam patterns. This paper addresses the problem of joint hybrid holographic beamforming and user scheduling under per-user minimum quality-of-service (QoS) constraints, a critical

Guijin Son, Jiwoo Hong, Hyunwoo Ko, James Thorne

Scaling pre-training compute has proven effective for achieving mulitlinguality, but does the same hold for test-time scaling? In this work, we introduce MCLM, a multilingual math benchmark featuring competition-level problems in 55 languages. We test three test-time scaling methods-Outcome Reward Modeling (ORM), Process Reward Modeling (ORM), and Budget For

Rohit Saxena, Pasquale Minervini, Frank Keller

Generating accurate and concise textual summaries from multimodal documents is challenging, especially when dealing with visually complex content like scientific posters. We introduce PosterSum, a novel benchmark to advance the development of vision-language models that can understand and summarize scientific posters into research paper abstracts. Our datase

Lintao Li, Xiye Hu, Zhubing Jia, William Huie

The integration of quantum computers and sensors into a quantum network opens a new frontier for quantum information science. We demonstrate high-fidelity entanglement between ytterbium-171 atoms -- the basis for state-of-the-art atomic quantum processors and optical atomic clocks -- and optical photons directly generated in the telecommunication wavelength

Edward Milsom, Ben Anson, Laurence Aitchison

We consider layerwise function-space learning rates, which measure the magnitude of the change in a neural network's output function in response to an update to a parameter tensor. This contrasts with traditional learning rates, which describe the magnitude of changes in parameter space. We develop efficient methods to measure and set function-space learning

Stefan Hegselmann, Georg von Arnim, Tillmann Rheude, Noel Kronenberg

Electronic Health Records (EHRs) offer considerable potential for clinical prediction, but their complexity and heterogeneity challenge traditional machine learning. Domain-specific EHR foundation models trained on unlabeled EHR data have shown improved predictive accuracy and generalization. However, their development is constrained by limited data access a

Karen Kang, Simona J. Miller, Katerina Chatziioannou, Deborah Ferguson

The spins of binary black holes measured with gravitational waves provide insights about the formation, evolution, and dynamics of these systems. However, interpreting these measurements-especially for heavy black holes-remains an open problem. While the imprint of spin during the inspiral phase, where the black holes are well-separated, is understood throug

Giulia Cusin, Cyril Pitrou, Martin Pijnenburg, Alberto Sesana

The astrophysical gravitational wave background in the nanohertz (nHz) band is expected to be primarily composed of the superposition of signals from binaries of supermassive black holes. The spatial discreteness of these sources introduces shot noise, which, in certain regimes, would overwhelm efforts to measure the anisotropy of the gravitational wave back

Sumit Kumar, Max Melching, Frank Ohme

The PE for GW merger events relies on a waveform model calibrated using numerical simulations. Within the Bayesian framework, this waveform model represents the GW signal produced during the merger and is crucial for estimating the likelihood function. However, these waveform models may possess systematic errors that can differ across the parameter space. Ad

Liuchuan Yu, Ching-I Huang, Hsueh-Cheng Wang, Lap-Fai Yu

Augmented reality (AR) games, particularly those designed for head-mounted displays, have grown increasingly prevalent. However, most existing systems depend on pre-scanned, static environments and rely heavily on continuous tracking or marker-based solutions, which limit adaptability in dynamic physical spaces. This is particularly problematic for AR headse

Cyril Koenig, Enrico Zelioli, Luca Benini

Embedded heterogeneous systems-on-chip (SoCs) rely on domain-specific hardware accelerators to improve performance and energy efficiency. In particular, programmable multi-core accelerators feature a cluster of processing elements and tightly coupled scratchpad memories to balance performance, energy efficiency, and flexibility. In embedded systems running a

Yoeri Poels, Cristina Venturini, Alessandro Pau, Olivier Sauter

Maximizing fusion performance in tokamaks relies on high energy confinement, often achieved through distinct operating regimes. The automated labeling of these confinement states is crucial to enable large-scale analyses or for real-time control applications. While this task becomes difficult to automate near state transitions or in marginal scenarios, much

Luca Pezzè, Augusto Smerzi

Recent years have witnessed a growing interest in understating the limitations imposed by quantum noise in precision measurements and devising techniques to reduce it. The attention is currently turning to the simultaneously estimation of several parameters of interest, driven by its promising potential across a wide range of sensing applications as well as

Marco Antonio Cardoso Alvarez, Leila Graef, Robert Brandenberger

Previous studies suggested that the back-reaction of super-Hubble cosmological fluctuations could lead to a dynamical relaxation of the cosmological constant. Moreover, this mechanism appears to be self-regulatory, potentially leading to an oscillatory behavior in the effective dark energy. Such an effect would occur in any cosmological model with super-Hubb

Petro Kolosov

Let $P(m, X, N)$ be an $m$-degree polynomial in $X\in\mathbb{R}$ having fixed non-negative integers $m$ and $N$. The polynomial $P(m, X, N)$ is derived from a rearrangement of Faulhaber's formula in the context of Knuth's work entitled "Johann Faulhaber and sums of powers". In this manuscript we discuss the approximation properties of polynomial $P(m,X,N)$.

Tanmay Parekh, Yuxuan Dong, Lucas Bandarkar, Artin Kim

Event Detection (ED) -- the task of identifying event mentions from natural language text -- is critical for enabling reasoning in highly specialized domains such as biomedicine, law, and epidemiology. Data generation has proven to be effective in broadening its utility to wider applications without requiring expensive expert annotations. However, when exist

Amanda Bertschinger, James Bagrow, Joshua Bongard

[RETRACTED]Data increasingly abounds, but distilling their underlying relationships down to something interpretable remains challenging. One approach is genetic programming, which `symbolically regresses' a data set down into an equation. However, symbolic regression (SR) faces the issue of requiring training from scratch for each new dataset. To generalize

Yangshijie Zhang

Deep neural networks (DNNs) have achieved remarkable success in the field of natural language processing (NLP), leading to widely recognized applications such as ChatGPT. However, the vulnerability of these models to adversarial attacks remains a significant concern. Unlike continuous domains like images, text exists in a discrete space, making even minor al

Andrei Chernov, Oleg Novitskij

Recent studies have shown that reducing symmetries in neural networks enhances linear mode connectivity between networks without requiring parameter space alignment, leading to improved performance in linearly interpolated neural networks. However, in practical applications, neural network interpolation is rarely used; instead, ensembles of networks are more

Taeyoun Kim, Jacob Springer, Aditi Raghunathan, Maarten Sap

In retrieval augmented generation (RAG) systems, each individual component -- the LLM, embedder, and corpus -- could introduce biases in the form of skews towards outputting certain perspectives or identities. In this work, we study the conflict between biases of each component and their relationship to the overall bias of the RAG system, which we call bias

Ali Amhaz, Shreya Khisa, Mohamed Elhattab, Chadi Assi

This study investigates a downlink rate-splitting multiple access (RSMA) scenario in which multiple base stations (BSs), employing a coordinated multi-point (CoMP) transmission scheme, serve users equipped with movable antenna (MA) technology. Unlike traditional fixed-position antennas (FPAs), which are subject to random variations in wireless channels, MAs

Adnan A. E. Hajomer, Ivan Derkach, Vladyslav C. Usenko, Ulrik L. Andersen

Integrating quantum key distribution (QKD) with classical data transmission over the same fiber is crucial for scalable quantum-secured communication. However, noise from classical channels limits QKD distance. We demonstrate the longest-distance continuous-variable QKD (CV-QKD) over 120 km (20 dB loss) in the asymptotic regime, and over 100 km (17 dB loss)

Alon Albalak, Duy Phung, Nathan Lile, Rafael Rafailov

Increasing interest in reasoning models has led math to become a prominent testing ground for algorithmic and methodological improvements. However, existing open math datasets either contain a small collection of high-quality, human-written problems or a large corpus of machine-generated problems of uncertain quality, forcing researchers to choose between qu

N. Carey, L. Zanisi, S. Pamela, V. Gopakumar

The inclusion of high-fidelity simulations of SOL turbulence and transient MHD events such as ELMs in highly iterative applications remains computationally prohibitive, limiting their use in design and control workflows. Understanding these phenomena is vital, as they govern heat flux on plasma-facing components, influencing reactor performance and material

Juan Carlos Mongez, Maria José Pacifico, Mauricio Poletti

We prove the finiteness of ergodic measures of maximal entropy for partially hyperbolic diffeomorphisms where the center direction has a dominated decomposition into one dimensional bundle and there is a uniform lower bound for the absolute value of the Lyapunov exponents. As applications we prove finiteness for a class derived from Anosov partially hyperbol

Sasha Voitovych, Mahdi Haghifam, Idan Attias, Gintare Karolina Dziugaite

In this paper, we investigate the necessity of traceability for accurate learning in stochastic convex optimization (SCO) under $\ell_p$ geometries. Informally, we say a learning algorithm is $m$-traceable if, by analyzing its output, it is possible to identify at least $m$ of its training samples. Our main results uncover a fundamental tradeoff between trac

Dong-Ho Lee, Hyundong Cho, Jonathan May, Jay Pujara

Asking good questions is critical for comprehension and learning, yet evaluating and generating such questions remains a challenging problem. Prior work on inquisitive questions focuses on learner-generated, curiosity-driven queries and evaluates them using indirect metrics, such as salience or information gain, that do not directly capture a question's impa

Jiaqi Shang, Gabriel Kreiman, Haim Sompolinsky

Humans readily generalize abstract relations, such as recognizing "constant" in shape or color, whereas neural networks struggle, limiting their flexible reasoning. To investigate mechanisms underlying such generalization, we introduce SimplifiedRPM, a novel benchmark for systematically evaluating abstract relational reasoning, addressing limitations in prio

Zhi-Lei She, An-Ke Lei, Dai-Mei Zhou, Larissa V. Bravina

The coalescence production of sexaquark, a hypothetical stable state with quark content $(uuddss)$, is investigated by the parton and hadron cascade model PACIAE in $pp$ collisions at $\sqrt s = 7$ TeV. In this work, the compact sexaquark bound state of three diquarks is formed in the final partonic state by a two-step approach, which involves ``diquark" for

Qiuming Zhao, Guangzhi Sun, Chao Zhang

Language diversity presents a significant challenge in speech-to-text (S2T) tasks, such as automatic speech recognition and translation. Traditional multi-lingual multi-task training approaches aim to address this by jointly optimising multiple speech recognition and translation tasks across various languages. While models like Whisper, built on these strate

Adhish Rele

We extend the study of Hall algebras and edge contractions by generalizing Yiqiang Li's work to contraction along vertices with multiple edges. Using the edge contractions, we establish new embeddings among Hall algebras in this broader setting. Our results demonstrate that these embeddings preserve key algebraic structures, including Hopf algebra operations

João Helis Bernardo, Daniel Alencar da Costa, Filipe Roseiro Cogo, Sérgio Queiróz de Medeiros

Continuous Integration (CI) is a cornerstone of modern software development. However, while widely adopted in traditional software projects, applying CI practices to Machine Learning (ML) projects presents distinctive characteristics. For example, our previous work revealed that ML projects often experience longer build durations and lower test coverage rate

Chong Cheng, Gaochao Song, Yiyang Yao, Qinzheng Zhou

This paper investigates an open research challenge of reconstructing high-quality, large 3D open scenes from images. It is observed existing methods have various limitations, such as requiring precise camera poses for input and dense viewpoints for supervision. To perform effective and efficient 3D scene reconstruction, we propose a novel graph-guided 3D sce

Allison R. Pessoa, Leonardo de S. Menezes, Anderson M. Amaral

Trivalent lanthanide ion-doped nanoparticles are widely employed as nanoscale thermometers, driving rapid advancements in real-world applications. When the Luminescence Intensity Ratio (LIR) technique is used, these thermometric systems typically require a calibration process to obtain macroscopic calibration parameters. However, despite extensive studies fr

E. Franco, J. J. L. Velázquez

The goal of this paper is to understand if the property of adaptation, which is a typical property of many biochemical systems, can be achieved only by biological systems that actively consume energy or if it can be achieved also by passive systems. We prove that, unless the conserved quantities of a signalling system satisfy a very specific factorization as

Dmitry Gorbunov, Dmitry Kalashnikov, George Krugan

We examine the sterile neutrino dark matter production in the primordial plasma with lepton asymmetry unequally distributed over different neutrino flavors. We argue that with the specific flavor fractions, one can mitigate limits from the Big Bang Nucleosynthesis on the sterile-active neutrino mixing angle and sterile neutrino mass. It happens due to cancel

Jun Liu

Convergence analysis of Nesterov's accelerated gradient method has attracted significant attention over the past decades. While extensive work has explored its theoretical properties and elucidated the intuition behind its acceleration, a simple and direct proof of its convergence rates is still lacking. We provide a concise Lyapunov analysis of the converge

Stella Dumenčić, Luka Lanča, Karlo Jakac, Stefan Ivić

Search and rescue (SAR) missions require reliable search methods to locate survivors, especially in challenging or inaccessible environments. This is why introducing unmanned aerial vehicles (UAVs) can be of great help to enhance the efficiency of SAR missions while simultaneously increasing the safety of everyone involved in the mission. Motivated by this,

Emiliano Seri, Marcello Petitta, Chryssoula Papaioannou, Nikolaos Katsoulas

The integration of photovoltaic (PV) systems into greenhouses not only optimizes land use but also enhances sustainable agricultural practices by enabling dual benefits of food production and renewable energy generation. However, accurate prediction of internal environmental conditions is crucial to ensure optimal crop growth while maximizing energy producti

Fulvio Melia

JWST has made several surprising discoveries, underscored by the `too early' appearance of well-formed galaxies and supermassive black holes. It recently also uncovered a compact galaxy (JWST-ER1g) associated with a complete Einstein ring (JWST-ER1r) at photometric redshift $z_l=1.94^{+0.13}_{-0.17}$, produced by a lensed galaxy at $z_s=2.98^{+0.42}_{-0.47}$

Simone Drago, Marco Mussi, Alberto Maria Metelli

In this work, we provide a refined analysis of the UCBVI algorithm (Azar et al., 2017), improving both the bonus terms and the regret analysis. Additionally, we compare our version of UCBVI with both its original version and the state-of-the-art MVP algorithm. Our empirical validation demonstrates that improving the multiplicative constants in the bounds has

A. Rubio, S. Rodríguez-Aparicio, J. M. Montanero, M. G. Cabezas

We present a novel microfluidic method to produce quasi-monodisperse bubbles with diameters from tens to very few microns. A gaseous rivulet flows over the shallow groove printed on a T-junction exit channel. The triple contact line delimiting the rivulet is pinned to the groove edges. The rivulet breaks up into bubbles much smaller than the exit channel. Wh

John Preskill

Today's Noisy Intermediate-Scale Quantum (NISQ) computers have scientific value, but quantum machines with broad practical value must be protected against noise using quantum error correction and fault-tolerant protocols. Recent studies of quantum error correction on actual hardware are opening a new era of quantum information processing. Error-corrected com

Louise Kimpton, James Salter, Xiaoyu Xiong, Peter Challenor

Decision making often uses complex computer codes run at the exa-scale (10e18 flops). Such computer codes or models are often run in a hierarchy of different levels of fidelity ranging from the basic to the very sophisticated. The top levels in this hierarchy are expensive to run, limiting the number of possible runs. To make use of runs over all levels, and

Jikang Deng, Hui Zhou, Mohamed-Slim Alouini

To achieve global coverage and ubiquitous connectivity, the non-terrestrial network (NTN) has been regarded as a key enabler in the sixth generation (6G) network, which includes uncrewed aerial vehicles (UAVs), high-altitude platforms (HAPs), and satellites. Since the unique characteristics of various NTN platforms strongly affect their implementation and le

Marcel Hudiani, Claudio Landim, Sunder Sethuraman

We derive a singular diffusion limit for the position of a tagged particle in zero range interacting particle processes on a one dimensional torus with a Sinai-type random environment via two steps. In the first step, a regularization is introduced by averaging the random environment over an $\varepsilon N$-neighborhood. With respect to such an environment,

Toheeb Aduramomi Jimoh, Tabea De Wille, Nikola S. Nikolov

Natural Language Processing (NLP) is becoming a dominant subset of artificial intelligence as the need to help machines understand human language looks indispensable. Several NLP applications are ubiquitous, partly due to the myriad of datasets being churned out daily through mediums like social networking sites. However, the growing development has not been

Tianrui Zhu, Shiyi Zhang, Jiawei Shao, Yansong Tang

Background consistency remains a significant challenge in image editing tasks. Despite extensive developments, existing works still face a trade-off between maintaining similarity to the original image and generating content that aligns with the target. Here, we propose KV-Edit, a training-free approach that uses KV cache in DiTs to maintain background consi

Julien Mellet, Fabio Ruggiero, Vincenzo Lippiello

Humans process significantly more information through the sense of touch than through vision. Consequently, haptics for telemanipulation is poised to become essential in the coming years, as it offers operators an additional sensory channel crucial for interpretation in extreme conditions. However, current haptic device setups are either difficult to access

Han-Jia Ye, Si-Yang Liu, Wei-Lun Chao

Tabular datasets are inherently heterogeneous, presenting significant challenges for developing pre-trained foundation models. The recently introduced transformer-based Tabular Prior-data Fitted Network v2 (TabPFN v2) achieves unprecedented in-context learning performance across diverse downstream datasets, marking a pivotal advancement in tabular foundation

Orhun Utku Aydin, Alexander Koch, Adam Hilbert, Jana Rieger

Despite the potential of synthetic medical data for augmenting and improving the generalizability of deep learning models, memorization in generative models can lead to unintended leakage of sensitive patient information and limit model utility. Thus, the use of memorizing generative models in the medical domain can jeopardize patient privacy. We propose a f

Alexander Hammerl, Ravi Seshadri, Thomas Kjær Rasmussen, Otto Anker Nielsen

This paper analyzes the time-dependent relationship between the mean and variance of travel time on a single corridor under rush hour like congestion patterns. To model this phenomenon, we apply the LWR ((Lighthill & Whitham, 1955), (Richards, 1956)) theory on a homogenous freeway with a discontinuous bottleneck at its downstream end, assuming a uni-modal de

André V. Duarte, Xuandong Zhao, Arlindo L. Oliveira, Lei Li

How can we verify whether copyrighted content was used to train a large vision-language model (VLM) without direct access to its training data? Motivated by the hypothesis that a VLM is able to recognize images from its training corpus, we propose DIS-CO, a novel approach to infer the inclusion of copyrighted content during the model's development. By repeat

Israrul H Hashmi, Rahul Karmakar, Marripelli Maniteja, Kumar Ayush

Lennard-Jones (LJ) fluids serve as an important theoretical framework for understanding molecular interactions. Binary LJ fluids, where two distinct species of particles interact based on the LJ potential, exhibit rich phase behavior and provide valuable insights of complex fluid mixtures. Here we report the construction and utility of an artificial intellig

Rosie Zhao, Tian Qin, David Alvarez-Melis, Sham Kakade

Language models famously improve under a smooth scaling law, but some specific capabilities exhibit sudden breakthroughs in performance. Advocates of "emergence" view these capabilities as unlocked at a specific scale, but others attribute breakthroughs to superficial metric thresholding effects. We propose that breakthroughs are instead driven by continuous

A. H. Olabintan, E. Lawan, T. A Oluwadare

Earth's net radiation, sometimes referred to as net flux, is the balance between incoming and outgoing solar radiation energy from the atmosphere. This study utilised five years of data on tropospheric variables (solar radiation, air temperature, surface temperature, maximum temperature, minimum temperature, and dew point temperature), from January 2007 to D

Yihong Liu, Runsheng Chen, Lea Hirlimann, Ahmad Dawar Hakimi

In large language models (LLMs), certain \emph{neurons} can store distinct pieces of knowledge learned during pretraining. While factual knowledge typically appears as a combination of \emph{relations} and \emph{entities}, it remains unclear whether some neurons focus on a relation itself -- independent of any entity. We hypothesize such neurons \emph{detect

Thomas Lavigne, Stéphane Urcun, Bérengère Fromy, Audrey Josset-Lamaugarny

Research on human skin anatomy reveals its complex multi-scale, multi-phase nature, with up to 70% of its composition being bounded and free water. Fluid movement plays a key role in the skin's mechanical and biological responses, influencing its time-dependent behavior and nutrient transport. Poroelastic modeling is a promising approach for studying skin dy

Emma Dodd, Tadafumi Matsuno, Amina Helmi, Eduardo Balbinot

The local stellar halo of the Milky Way is known to contain the debris from accreted dwarf galaxies and globular clusters, in the form of stellar streams and over-densities in the space of orbital properties (e.g. integrals of motion). While several over-densities have been uncovered and characterised dynamically using Gaia data, their nature is not always c

Karan Samel, Nitish Sontakke, Irfan Essa

Instructional videos provide a convenient modality to learn new tasks (ex. cooking a recipe, or assembling furniture). A viewer will want to find a corresponding video that reflects both the overall task they are interested in as well as contains the relevant steps they need to carry out the task. To perform this, an instructional video model should be capab

Raj Korpan

As robots take on caregiving roles, ensuring equitable and unbiased interactions with diverse populations is critical. Although Large Language Models (LLMs) serve as key components in shaping robotic behavior, speech, and decision-making, these models may encode and propagate societal biases, leading to disparities in care based on demographic factors. This

Christine Gaßner

Most ideas about what an algorithm is are very similar. Basic operations are used for transforming objects. The evaluation of internal and external states by relations has impact on the further process. A more precise definition can lead to a model of abstract computation over an arbitrary first-order structure. Formally, the algorithms can be determined by

A. Deltuva, D. Jurčiukonis, D. Likandrovas, J. Torres Fernandez

We test microscopic global optical potential in three-body calculations of deuteron-nucleus scattering. We solve Faddeev-type equations for three-body transition operators. We calculate differential cross section and analyzing power for the deuteron elastic scattering and breakup in collisions with ${}^{12}$C, ${}^{16}$O and ${}^{24}$Mg nuclei, and find a re

Bohan Zhang, Yixin Wang, Paramveer S. Dhillon

This paper introduces a novel causal framework for multi-stage decision-making in natural language action spaces where outcomes are only observed after a sequence of actions. While recent approaches like Proximal Policy Optimization (PPO) can handle such delayed-reward settings in high-dimensional action spaces, they typically require multiple models (policy

Alex D. Richardson, Kaicheng Zhang, Lucas Beerens, Dongdong Chen

The proliferation of text-to-image diffusion models has raised significant privacy and security concerns, particularly regarding the generation of copyrighted or harmful images. In response, concept erasure (defense) methods have been developed to "unlearn" specific concepts through post-hoc finetuning. However, recent concept restoration (attack) methods ha

Polina Kutsevol, Onur Ayan, Nikolaos Pappas, Wolfgang Kellerer

This work explores employing the concept of goal-oriented (GO) semantic communication for real-time monitoring and control. Generally, GO communication advocates for the deep integration of application targets into the network design. We consider CPS and IoT applications where sensors generate a tremendous amount of network traffic toward monitors or control

Minyeong Hwang, Ziseok Lee, Kwang-Soo Kim, Kyungsu Kim

Linker generation is critical in drug discovery applications such as lead optimization and PROTAC design, where molecular fragments are assembled into diverse drug candidates via molecular linker. Existing methods fall into point cloud-free and point cloud-aware categories based on their use of fragments' 3D poses alongside their topologies in sampling the l

Gabrielle O'Brien

Scientists across disciplines write code for critical activities like data collection and generation, statistical modeling, and visualization. As large language models that can generate code have become widely available, scientists may increasingly use these models during research software development. We investigate the characteristics of scientists who are

Francesco Vona, Maximilian Warsinke, Tanja Kojic, Jan-Niklas Voit-Antons

The integration of Digital Twins with Extended Reality technologies, such as Virtual Reality and Augmented Reality, is transforming industries by enabling more immersive, interactive experiences and enhancing real time decision making. User centered evaluations are crucial for aligning XR enhanced DT systems with user expectations, enhancing acceptance and u

João Paulo Esper, Luciano de S. Fraga, Aline C. Viana, Kleber Vieira Cardoso

Next-generation touristic services will rely on the advanced mobile networks' high bandwidth and low latency and the Multi-access Edge Computing (MEC) paradigm to provide fully immersive mobile experiences. As an integral part of travel planning systems, recommendation algorithms devise personalized tour itineraries for individual users considering the popul

Valeria Pantè, David Axelrod, Alessandro Flammini, Filippo Menczer

Social media platforms have become key tools for coordinated influence operations, enabling state actors to manipulate public opinion through strategic, collective actions. While previous research has suggested collaboration between states, such research failed to leverage state-of-the-art coordination indicators or control datasets. In this study, we invest

Giulia Ricciardi, Natascia Vignaroli, Francesco Vissani

We analyze the cross section for inverse beta decay, focusing on the moderate energies (a few MeV to hundreds of MeV) relevant for reactor and supernova neutrinos. We discuss the updated evaluations of values and uncertainties in the cross section, and the effect of second-class currents. The estimate of theoretical precision is important for current and fut