May 2025 arXiv papers — page 64

Xiao-Dong Cheng, Zhen-Lu Weng, Ying-Ying Fan, Ru-Min Wang

In this paper, we investigate the CP asymmetries ${\mathcal A}_{CP}^{pm}$ and ${\mathcal A}_{CP}^{mp}$ in $B_{c}^{\pm}\rightarrow B^{\pm} K^{0}+B^{\pm} {\bar{K}}^0\rightarrow B^{\pm} \pi^{\pm} e^{\mp} \nu_{e}$ and $B_{c}^{\pm}\rightarrow B^{\pm} K^{0}+B^{\pm} {\bar{K}}^0\rightarrow B^{\pm} \pi^{\mp} e^{\pm} \nu_{e}$ decays, both of them consist of three part

Eric Tillman Bill, Cristian Perez Jensen, Sotiris Anagnostidis, Dimitri von Rütte

Denoising diffusion models exhibit remarkable generative capabilities, but remain challenging to train due to their inherent stochasticity, where high-variance gradient estimates lead to slow convergence. Previous works have shown that magnitude preservation helps with stabilizing training in the U-net architecture. This work explores whether this effect ext

Seunguk Yu, Juhwan Choi, Youngbin Kim

Despite the recent strides in large language models, studies have underscored the existence of social biases within these systems. In this paper, we delve into the validation and comparison of the ethical biases of LLMs concerning globally discussed and potentially sensitive topics, hypothesizing that these biases may arise from language-specific distinction

Aakash Sen Sharma, Debdeep Sanyal, Priyansh Srivastava, Sundar Atreya H.

Large language models (LLMs) commonly risk copyright infringement by reproducing protected content verbatim or with insufficient transformative modifications, posing significant ethical, legal, and practical concerns. Current inference-time safeguards predominantly rely on restrictive refusal-based filters, often compromising the practical utility of these m

Renyuan Li, Zhibo Liang, Haichuan Zhang, Tianyu Shi

Recent breakthroughs in text-to-speech (TTS) voice cloning have raised serious privacy concerns, allowing highly accurate vocal identity replication from just a few seconds of reference audio, while retaining the speaker's vocal authenticity. In this paper, we introduce CloneShield, a universal time-domain adversarial perturbation framework specifically desi

Sota Shimamura, Nen Saito, Shuji Ishihara

We numerically studied active Brownian particles with attractive interactions. Contrary to our intuition, the attractive force between particles disrupts the formation of a single cluster observed in motility-induced phase separation, giving rise to a multi-cluster state characterized by a power-law distribution of cluster sizes. Remarkably, the self-propuls

Hiroyuki Tajima, Eiji Nakano, Kei Iida

We show how a Fujita-Miyazawa-type three-body force emerges among three impurity atoms immersed in an atomic Bose-Einstein condensate near an interspecies Feshbach resonance. As a result of thermal average over excitations in the medium and impurities as well as expansion with respect to the impurity-medium and Feshbach resonance couplings, two superfluid ph

Nahyun Lee, Yeongseo Woo, Hyunwoo Ko, Guijin Son

Large language models often suffer from language confusion, a phenomenon in which responses are partially or entirely generated in unintended languages. This critically degrades the user experience, especially in low-resource settings. We hypothesize that this issue stems from limitations in conventional fine-tuning objectives, such as supervised learning, w

Brian Chmiel, Maxim Fishman, Ron Banner, Daniel Soudry

We demonstrate, for the first time, fully quantized training (FQT) of large language models (LLMs) using predominantly 4-bit floating-point (FP4) precision for weights, activations, and gradients on datasets up to 200 billion tokens. We extensively investigate key design choices for FP4, including block sizes, scaling formats, and rounding methods. Our analy

Hui Zhang, Dexiang Hong, Maoke Yang, Yutao Cheng

Graphic design plays a vital role in visual communication across advertising, marketing, and multimedia entertainment. Prior work has explored automated graphic design generation using diffusion models, aiming to streamline creative workflows and democratize design capabilities. However, complex graphic design scenarios require accurately adhering to design

Wen-Qi Li, Zhikai Zhang

In this paper, we establish a parabolic Harnack inequality for positive solutions of the $\phi$-heat equation and prove Gaussian upper and lower bounds for the $\phi$-heat kernel on weighted Riemannian manifolds under lower $N$-Ricci curvature bound with $\varepsilon$-range. Building on these results, we demonstrate: The $L^1_\phi$-Liouville theorem for $\ph

Zheng Chu, Huiming Fan, Jingchang Chen, Qianyu Wang

Although large language models (LLMs) have demonstrated remarkable reasoning capabilities, they still face challenges in knowledge-intensive multi-hop reasoning. Recent work explores iterative retrieval to address complex problems. However, the lack of intermediate guidance often results in inaccurate retrieval and flawed intermediate reasoning, leading to i

Yaping He, Jianfeng Cai, Qicong Hu, Peiqing Wang

To address the challenges posed by the large number of parameters in existing remote sensing image classification models, which hinder deployment on resource-constrained devices, this paper proposes a lightweight classification method based on knowledge distillation. Specifically, G-GhostNet is adopted as the backbone network, leveraging feature reuse to red

Vishwa Mohan Singh, Alberto Gaston Villagran Asiares, Luisa Sophie Schuhmacher, Kate Rendall

Diffusion Tensor Imaging (DTI) tractography offers detailed insights into the structural connectivity of the brain, but presents challenges in effective representation and interpretation in deep learning models. In this work, we propose a novel 2D representation of DTI tractography that encodes tract-level fractional anisotropy (FA) values into a 9x9 graysca

Yannic Maus, Janosch Ruff

We analyse the performance of simple distributed colouring algorithms under the assumption that the input graph is a hyperbolic random graph (HRG), a generative model capturing key properties of real-world networks such as power-law degree distributions and large clustering coefficients. Motivated by the shift from worst-case analysis to more realistic netwo

Yongheng Zhang, Xu Liu, Ruoxi Zhou, Qiguang Chen

Investigating hallucination issues in large language models (LLMs) within cross-lingual and cross-modal scenarios can greatly advance the large-scale deployment in real-world applications. Nevertheless, the current studies are limited to a single scenario, either cross-lingual or cross-modal, leaving a gap in the exploration of hallucinations in the joint cr

Boyan Gao, Xin Wang, Yibo Yang, David Clifton

Large Language Models (LLMs) have demonstrated remarkable performance in real-world applications. However, adapting LLMs to novel tasks via fine-tuning often requires substantial training data and computational resources that are impractical in few-shot scenarios. Existing approaches, such as in-context learning and Parameter-Efficient Fine-Tuning (PEFT), fa

Sai Li, Yihao Zhang, Albert Lee, Zheng Zhu

Physics-inspired computing paradigms, such as Ising machines, are emerging as promising hardware alternatives to traditional von Neumann architectures for tackling computationally intensive combinatorial optimization problems (COPs). While quantum, optical, and electronic devices have garnered significant attention for their potential in realizing Ising mach

Karn Tiwari, Niladri Dutta, N M Anoop Krishnan, Prathosh A P

Neural operators have emerged as powerful data-driven frameworks for solving Partial Differential Equations (PDEs), offering significant speedups over numerical methods. However, existing neural operators struggle with scalability in high-dimensional spaces, incur high computational costs, and face challenges in capturing continuous and long-range dependenci

Iddo Yosha, Dorin Shteyman, Yossi Adi

Spoken language conveys meaning not only through words but also through intonation, emotion, and emphasis. Sentence stress, the emphasis placed on specific words within a sentence, is crucial for conveying speaker intent and has been extensively studied in linguistics. In this work, we introduce WHISTRESS, an alignment-free approach for enhancing transcripti

Sergey Samsonov, Marina Sheshukova, Eric Moulines, Alexey Naumov

In this paper we derive non-asymptotic Berry-Esseen bounds for Polyak-Ruppert averaged iterates of the Linear Stochastic Approximation (LSA) algorithm driven by the Markovian noise. Our analysis yields $\mathcal{O}(n^{-1/4})$ convergence rates to the Gaussian limit in the Kolmogorov distance. We further establish the non-asymptotic validity of a multiplier b

Vaishali Dhanoa, Anton Wolter, Gabriela Molina León, Hans-Jörg Schulz

Autonomous agents powered by Large Language Models are transforming AI, creating an imperative for the visualization field to embrace agentic frameworks. However, our field's focus on a human in the sensemaking loop raises critical questions about autonomy, delegation, and coordination for such \textit{agentic visualization} that preserve human agency while

Yeyuan Wang, Dehong Gao, Rujiao Long, Lei Yi

Direct Preference Optimization (DPO) has gained significant attention for its simplicity and computational efficiency in aligning large language models (LLMs). Recent advancements have extended DPO to multimodal scenarios, achieving strong performance. However, traditional DPO relies on binary preference optimization, rewarding or penalizing entire responses

Kun Xiang, Heng Li, Terry Jingchen Zhang, Yinya Huang

We present SeePhys, a large-scale multimodal benchmark for LLM reasoning grounded in physics questions ranging from middle school to PhD qualifying exams. The benchmark covers 7 fundamental domains spanning the physics discipline, incorporating 21 categories of highly heterogeneous diagrams. In contrast to prior works where visual elements mainly serve auxil

Vignesh Kottayam Viswanathan, Akash Patel, Mario Alberto Valdes Saucedo, Sumeet Satpute

In this work, we introduce SPADE, a path planning framework designed for autonomous navigation in dynamic environments using 3D scene graphs. SPADE combines hierarchical path planning with local geometric awareness to enable collision-free movement in dynamic scenes. The framework bifurcates the planning problem into two: (a) solving the sparse abstract glob

Xichen Ye, Yifan Wu, Weizhong Zhang, Cheng Jin

The Influence Function (IF) is a widely used technique for assessing the impact of individual training samples on model predictions. However, existing IF methods often fail to provide reliable influence estimates in deep neural networks, particularly when applied to noisy training data. This issue does not stem from inaccuracies in parameter change estimatio

Sabrina Patania, Luca Annese, Cansu Koyuturk, Azzurra Ruggeri

Large Language Models (LLMs) have demonstrated remarkable capabilities in processing extensive offline datasets. However, they often face challenges in acquiring and integrating complex, knowledge online. Traditional AI training paradigms, predominantly based on supervised learning or reinforcement learning, mirror a 'Piagetian' model of independent explorat

Qiong Li, Chao Fang, Longwei Huang, Jun Lin

While posit format offers superior dynamic range and accuracy for transprecision computing, its adoption in RISC-V processors is hindered by the lack of a unified solution for lightweight, precision-scalable, and IEEE-754 arithmetic compatible hardware implementation. To address these challenges, we enhance RISC-V processors by 1) integrating dedicated posit

Runliang Niu, Jinglong Ji, Yi Chang, Qi Wang

The rapid progress of large language models (LLMs) has sparked growing interest in building Artificial General Intelligence (AGI) within Graphical User Interface (GUI) environments. However, existing GUI agents based on LLMs or vision-language models (VLMs) often fail to generalize to novel environments and rely heavily on manually curated, diverse datasets.

Chuming Shen, Wei Wei, Xiaoye Qu, Yu Cheng

DeepSeek-R1 has demonstrated powerful reasoning capabilities in the text domain through stable reinforcement learning (RL). Recently, in the multimodal domain, works have begun to directly apply RL to generate R1-like free-form reasoning for Visual Question Answering (VQA) tasks. However, multimodal tasks share an intrinsically different nature from textual

Binh H. Ho, Long Nguyen Chi, TrungTin Nguyen, Binh T. Nguyen

Model-based clustering integrated with variable selection is a powerful tool for uncovering latent structures within complex data. However, its effectiveness is often hindered by challenges such as identifying relevant variables that define heterogeneous subgroups and handling data that are missing not at random, a prevalent issue in fields like transcriptom

Yang Zhang, Wenxin Xu, Xiaoyan Zhao, Wenjie Wang

Large Language Models (LLMs) have demonstrated impressive reasoning capabilities in complex problem-solving tasks, sparking growing interest in their application to preference reasoning in recommendation systems. Existing methods typically rely on fine-tuning with explicit chain-of-thought (CoT) data. However, these methods face significant practical limitat

Benjamin Clavié, Florian Brand

Recent advancements in Large Vision-Language Models (VLMs), have greatly enhanced their capability to jointly process text and images. However, despite extensive benchmarks evaluating visual comprehension (e.g., diagrams, color schemes, OCR tasks...), there is limited assessment of VLMs' ability to read and reason about text-rich images effectively. To fill

Haotian Si, Changhua Pei, Jianhui Li, Dan Pei

Recent advances in lightweight time series forecasting models suggest the inherent simplicity of time series forecasting tasks. In this paper, we present CMoS, a super-lightweight time series forecasting model. Instead of learning the embedding of the shapes, CMoS directly models the spatial correlations between different time series chunks. Additionally, we

Xuejie Liu, Anji Liu, Guy Van den Broeck, Yitao Liang

Masked generative models (MGMs) have emerged as a powerful framework for image synthesis, combining parallel decoding with strong bidirectional context modeling. However, generating high-quality samples typically requires many iterative decoding steps, resulting in high inference costs. A straightforward way to speed up generation is by decoding more tokens

Ajai Choudhry

Euler had considered the problem of finding three integers whose sum, product, and also the sum of the products of the integers, taken two at a time, are all perfect squares. Euler's methods of solving the problem lead to parametric solutions in terms of polynomials of high degrees and his numerical solutions consisted of very large integers. We obtain, by a

Itamar Harel, Yonathan Wolanowsky, Gal Vardi, Nathan Srebro

We analyze the generalization gap (gap between the training and test errors) when training a potentially over-parametrized model using a Markovian stochastic training algorithm, initialized from some distribution $\theta_0 \sim p_0$. We focus on Langevin dynamics with a positive temperature $\beta^{-1}$, i.e. gradient descent on a training loss $L$ with infi

Chen Tessler, Yifeng Jiang, Erwin Coumans, Zhengyi Luo

We tackle the challenges of synthesizing versatile, physically simulated human motions for full-body object manipulation. Unlike prior methods that are focused on detailed motion tracking, trajectory following, or teleoperation, our framework enables users to specify versatile high-level objectives such as target object poses or body poses. To achieve this,

Lei Guo, Chenlong Song, Feng Guo, Xiaohui Han

Non-overlapping Cross-domain Sequential Recommendation (NCSR) is the task that focuses on domain knowledge transfer without overlapping entities. Compared with traditional Cross-domain Sequential Recommendation (CSR), NCSR poses several challenges: 1) NCSR methods often rely on explicit item IDs, overlooking semantic information among entities. 2) Existing C

Yifeng Xu, Zhenliang He, Meina Kan, Shiguang Shan

Visual generation and understanding are two deeply interconnected aspects of human intelligence, yet they have been traditionally treated as separate tasks in machine learning. In this paper, we propose Jodi, a diffusion framework that unifies visual generation and understanding by jointly modeling the image domain and multiple label domains. Specifically, J

Purba Mukherjee, Anjan A Sen

We perform a model-independent reconstruction of the cosmic distances using the Multi-Task Gaussian Process (MTGP) framework as well as knot-based spline techniques with DESI-DR2 BAO and DES-SN5YR datasets. We calibrate the comoving sound horizon at the baryon drag epoch $r_d$ to the Planck value, ensuring consistency with early-universe physics. With the re

Bo-hyun Kwon

In this paper, we define the \textit{normal form} and \textit{normal coordinate} of a rational 3-tangle $T$ with respect to $\partial E_1$, where $E_1$ is the fixed two punctured disk in $\Sigma_{0,6}$. Among all normal coordinates of $T$ with respect to $\partial E_1$, we investigate the collection of \textit{minimal} normal coordinates of $T$. We show that

Ruiyang Xia, Dawei Zhou, Decheng Liu, Lin Yuan

Face swapping, recognized as a privacy and security concern, has prompted considerable defensive research. With the advancements in AI-generated content, the discrepancies between the real and swapped faces have become nuanced. Considering the difficulty of forged traces detection, we shift the focus to the face swapping purpose and proactively embed elabora

Tuan Van Vo, Tan Quang Nguyen, Khang Minh Nguyen, Duy Ho Minh Nguyen

Vision-Language-Action (VLA) models have gained much attention from the research community thanks to their strength in translating multimodal observations with linguistic instructions into robotic actions. Despite their recent advancements, VLAs often overlook the explicit reasoning and only learn the functional input-action mappings, omitting these crucial

L. H. Wei, H. J. Xing, L. B. Fu, H. D. Liu

Quantum Fisher Information (QFI) is a fundamental quantity in quantum parameter estimation theory, characterizing the ultimate precision bound of parameter estimation. In this work, we investigate QFI for quantum states in non-Hermitian systems. By employing the projected Hilbert space method and spectral decomposition, we derive an explicit expression for t

Farzaneh Derakhshan, Jan Hoffmann

This volume contains the proceedings of PLACES 2025, the 16th edition of the Workshop on Programming Language Approaches to Concurrency and Communication-cEntric Software. The workshop is scheduled to take place in Hamilton, Canada, on May 4, 2025, as a satellite event of ETAPS, the European Joint Conferences on Theory and Practice of Software. PLACES offers

Suyang Hu, Xiaoxu Lyu, Peihu Duan, Dawei Shi

This article investigates the problem of data-driven state estimation for linear systems with both unknown system dynamics and noise covariances. We propose an Autocovariance Least-squares-based Data-driven Kalman Filter (ADKF), which provides a unified framework for simultaneous system identification and state estimation by utilizing pre-collected input-out

Muye Huang, Lingling Zhang, Jie Ma, Han Lai

Charts are high-density visualization carriers for complex data, serving as a crucial medium for information extraction and analysis. Automated chart understanding poses significant challenges to existing multimodal large language models (MLLMs) due to the need for precise and complex visual reasoning. Current step-by-step reasoning models primarily focus on

Anders Björn, Jana Björn, Sylvester Eriksson-Bique, Xiaodan Zhou

We study uniqueness of $p$-harmonic Green functions in domains $\Omega$ in a complete metric space equipped with a doubling measure supporting a $p$-Poincar\'e inequality, with $1<p<\infty$. For bounded domains in unweighted $\mathbf{R}^n$, the uniqueness was shown for the $p$-Laplace operator $\Delta_p$ and all $p$ by Kichenassamy--V\'eron (Math. Ann. 275 (

Rui Li, Jing Long, Muge Qi, Heming Xia

To facilitate robust and trustworthy deployment of large language models (LLMs), it is essential to quantify the reliability of their generations through uncertainty estimation. While recent efforts have made significant advancements by leveraging the internal logic and linguistic features of LLMs to estimate uncertainty scores, our empirical analysis highli

Peter L. Guo, Mingyang Kang, Jiaji Liu

For a skew shape $\lambda/\mu$, we define the hybrid Grothendieck polynomial $${G}_{\lambda/\mu}(\textbf{x};\textbf{t};\textbf{w}) =\sum_{T\in \mathrm{SVRPP}(\lambda/\mu)} \textbf{x}^{\mathrm{ircont}(T)}\textbf{t}^{\mathrm{ceq} (T)}\textbf{w}^{\mathrm{ex}(T)}$$ as a weight generating function over set-valued reverse plane partitions of shape $\lambda/\mu$. I

Asad Ali, M. I. Hussain, Saif Al-Kuwari, M. T. Rahim

Quantum coherence and phase transitions are studied in a finite one-dimensional Bose--Hubbard model using exact diagonalization under thermal fluctuations, a Stark potential, and disorder. The condensate fraction, superfluid fraction, visibility, number fluctuations, and the $\ell_1$-norm of coherence are computed to characterize the Mott insulator--superflu

Fan Zhang, Lan Yin

Recent repoet on the formation of two-dimensional Bose-Einstein condensates (BECs) of spinning g-wave molecules is surprise. Here we study quantum phase transition in the quasi-2D atomic Bose gas with a g-wave Feshbach resonance, and show that there are two phase transitions in this system: from a phase with only a atomic Bose-Einstein condensate to a phase

Sebastian Stock, Michael Leuschel, Atif Mashkoor

When validating formal models, sizable effort goes into ensuring two types of properties: safety properties (nothing bad happens) and liveness properties (something good occurs eventually. Event-B supports checking safety properties all through the refinement chain. The same is not valid for liveness properties. Liveness properties are commonly validated wit

Dirk Tasche

Recalibration of binary probabilistic classifiers to a target prior probability is an important task in areas like credit risk management. However, recalibration of a classifier learned on a training dataset to a target on a test dataset in general is not a well-defined problem because there might be more than one way to transform the original posterior prob

Chang-Yin Huang, Yi Xie

The infrared (IR)/X-ray correlation of GX 339$-$4 is investigated based on a jet model with a modification by linking the magnetic field at the jet base to the accretion rate of the inner accretion flow though the the equilibrium between magnetic pressure at horizon and the ram pressure of the accretion flow. The IR flux is attributed to the synchrotron radi

Dong-Gang Wang, Bowei Zhang

Signatures of heavy particles during inflation are exponentially suppressed by the Boltzmann factor when the masses are far above the Hubble scale. In more realistic scenarios, however, scale-dependent features may change this conventional picture and boost the cosmological collider signals. In this paper, we compute cosmological correlators of the primordia

Jiashuo Chang, Zhengyi Li, Jianxun Lou, Zhen Qiu

Macro photography (MP) is a specialized field of photography that captures objects at an extremely close range, revealing tiny details. Although an accurate macro photography image quality assessment (MPIQA) metric can benefit macro photograph capturing, which is vital in some domains such as scientific research and medical applications, the lack of MPIQA da

Kacper Pryga, Bartlomiej Wiendlocha

This work presents a theoretical study of the electronic structure and transport properties of Ni-Au alloys, recently identified as excellent thermoelectric metals with a power factor significantly exceeding that of conventional semiconductor thermoelectrics. Using first-principles calculations based on the Korringa-Kohn-Rostoker method combined with the coh

Xin Ma, Yaohui Wang, Xinyuan Chen, Tien-Tsin Wong

Although diffusion models exhibit impressive generative capabilities, existing methods for stylized image generation based on these models often require textual inversion or fine-tuning with style images, which is time-consuming and limits the practical applicability of large-scale diffusion models. To address these challenges, we propose a novel stylized im

Ulrich Langer, Richard Löscher, Olaf Steinbach, Huidong Yang

We consider an abstract framework for the numerical solution of optimal control problems (OCPs) subject to partial differential equations (PDEs). Examples include not only the distributed control of elliptic PDEs such as the Poisson equation discussed in this paper in detail but also parabolic and hyperbolic equations. The approach covers the standard $L^2$

Roman Vashurin, Maiya Goloburda, Preslav Nakov, Maxim Panov

Large Language Models (LLMs) have become indispensable tools across various applications, making it more important than ever to ensure the quality and the trustworthiness of their outputs. This has led to growing interest in uncertainty quantification (UQ) methods for assessing the reliability of LLM outputs. Many existing UQ techniques rely on token probabi

Ignacio Mariano Andreozzi Pofcher, Joshua Ellul

Large Language Models (LLMs) are being used more and more for various coding tasks, including to help coders identify bugs and are a promising avenue to support coders in various tasks including vulnerability detection -- particularly given the flexibility of such generative AI models and tools. Yet for many tasks it may not be suitable to use LLMs, for whic

Ruilin Xu, Zongxuan Xie, Pengfei Chen

We present eACGM, a full-stack AI/ML system monitoring framework based on eBPF. eACGM collects real-time performance data from key hardware components, including the GPU and network communication layer, as well as from key software stacks such as CUDA, Python, and PyTorch, all without requiring any code instrumentation or modifications. Additionally, it leve

Chung I Lu, Julian Sester, Aijia Zhang

We propose a novel distributionally robust $Q$-learning algorithm for the non-tabular case accounting for continuous state spaces where the state transition of the underlying Markov decision process is subject to model uncertainty. The uncertainty is taken into account by considering the worst-case transition from a ball around a reference probability measur

Pedro Alonso, Tianrui Li, Chongshou Li

We challenge the common assumption that deeper decoder architectures always yield better performance in point cloud reconstruction. Our analysis reveals that, beyond a certain depth, increasing decoder complexity leads to overfitting and degraded generalization. Additionally, we propose a novel multi-head decoder architecture that exploits the inherent redun

Harethah Abu Shairah, Hasan Abed Al Kader Hammoud, Bernard Ghanem, George Turkiyyah

Large language models (LLMs) are typically aligned to refuse harmful instructions through safety fine-tuning. A recent attack, termed abliteration, identifies and suppresses the single latent direction most responsible for refusal behavior, thereby enabling models to generate harmful content. We propose a defense that fundamentally alters how models express

Xin Wang, Zhiqi Huang

Recent observational analyses have suggested possible evidence of hemisphere asymmetry in cosmological datasets. Parameterizations of this kind place observers in a privileged position-specifically on the plane that divides the two hemispheres. To quantify potential deviations from the cosmological principle without presuming a special location, we develop a

Zhuochen Liu, Rahul Jain, Quan Nguyen

Modern learning-based locomotion controllers typically rely on fully trainable deep neural networks with a large number of parameters. This paper studies a different design point for end-to-end control: whether effective quadruped locomotion can be achieved with a drastically reduced trainable parameter space. We present RANDomized POlicy Learning (RANDPOL),

Heiko Hoppe, Léo Baty, Louis Bouvier, Axel Parmentier

Reinforcement learning (RL) is increasingly applied to real-world problems involving complex and structured decisions, such as routing, scheduling, and assortment planning. These settings challenge standard RL algorithms, which struggle to scale, generalize, and exploit structure in the presence of combinatorial action spaces. We propose Structured Reinforce

Seokcheon Lee

Accurate measurements of fundamental cosmological parameters, especially the Hubble constant (H_0) and present-day matter density (\Omega_{m0}), are crucial for constraining dark energy (DE) evolution. We analyze the sensitivities of cosmological observables (H(z), D_L(z), E_{G}) to \Omega_{m0}, w_0, and w_an under different parametrizations. Our results sho

Mahdi Nikdan, Vincent Cohen-Addad, Dan Alistarh, Vahab Mirrokni

Effective data selection is critical for efficient training of modern Large Language Models (LLMs). This paper introduces Influence Distillation, a novel, mathematically-justified framework for data selection that employs second-order information to optimally weight training samples. By distilling each sample's influence on a target distribution, our method

Ahmad Moussavi, Peter Danchev, Arash Javan, Omid Hasanzadeh

This study explores in-depth the structure and properties of the so-called {\it strongly $\Delta$-clean rings}, that is a novel class of rings in which each ring element decomposes into a sum of a commuting idempotent and an element from the subset $\Delta(R)$. Here, $\Delta(R)$ stands for the extension of the Jacobson radical and is defined as the maximal s

Lu Wang, Xishuai Peng, S. Kevin Zhou

In recent years, more and more attention has been paid to the learning of 3D human representation. However, the complexity of lots of hand-defined human body constraints and the absence of supervision data limit that the existing works controllably and accurately represent the human body in views of semantics and representation ability. In this paper, we pro

Guangyu Zhu, Xidong Mu, Li Guo, Ao Huang

A novel movable-element simultaneously transmitting and reflecting surface (ME-STARS)-assisted near-field wideband communication framework is proposed. In particular, the position of each STARS element can be adjusted to combat the significant wideband beam squint issue in the near field instead of using costly true-time delay components. Four practical ME-S

Selçuk Topal

We introduce the MoveEVM Weakness Classification (MWC) system -- a dedicated vulnerability taxonomy for smart contracts built with Move and executed in EVM-compatible environments. While Move was originally designed to prevent common security flaws via linear resource types and strict ownership, its integration with EVM bytecode introduces novel hybrid vulne

Daniel Barzilai, Ohad Shamir

The widespread use of generative models has created a feedback loop, in which each generation of models is trained on data partially produced by its predecessors. This process has raised concerns about model collapse: A critical degradation in performance caused by repeated training on synthetic data. However, different analyses in the literature have reache

Christian Callaghan

This paper introduces Experiential Matrix Theory (EMT), a general theory of growth, employment, and technological change for the age of artificial intelligence (AI). EMT redefines utility as the alignment between production and an evolving, infinite-dimensional matrix of human experiential needs, thereby extending classical utility frameworks and integrating

Ryo Murakami, Kenji Miyatake, Ahmed Mohamed Ahmed Mahmoud, Hideki Yoshikawa

Anion-conductive polymer membranes have attracted considerable attention as solid electrolytes for alkaline fuel cells and electrolysis cells. Their hydroxide ion conductivity varies depending on factors such as the type and distribution of quaternary ammonium groups, as well as the structure and connectivity of hydrophilic and hydrophobic domains. In partic

Jingyuan Liu, Zeyu Zhang, Xuchuang Wang, Xutong Liu

Contextual multi-armed bandit is a fundamental learning framework for making a sequence of decisions, e.g., advertising recommendations for a sequence of arriving users. Recent works have shown that clustering these users based on the similarity of their learned preferences can accelerate the learning. However, prior work has primarily focused on the online

Mattias Nilsson-Takeuchi, Bharathram Ganapathisubramani

Many engineering and environmental surfaces exhibit spatial heterogeneity in the spanwise direction and encompass multiple surface length scales. When the dominant spanwise length scale is on the order of the largest flow scales (e.g., the boundary layer thickness or channel half-height, delta), localized delta-scale secondary flows can form. These secondary

Francesco Amicucci, Paola Leaci, Pia Astone, Sabrina D'Antonio

Continuous gravitational-wave signals (CWs), which are typically emitted by rapidly rotating neutron stars with non-axisymmetric deformations, represent particularly intriguing targets for the Advanced LIGO-Virgo-KAGRA detectors. These detectors operate within sensitivity bands that encompass more than half of the known pulsars in our galaxy existing in bina

Rawabi S. Al Qurashi, Maram M. Almnjomi, Teef L. Alghamdi, Amjad H. Almalki

Waste management is a critical global issue with significant environmental and public health implications. It has become more destructive during large-scale events such as the annual pilgrimage to Makkah, Saudi Arabia, one of the world's largest religious gatherings. This event's popularity has attracted millions worldwide, leading to significant and un-pred

Ke-Jung Chen

Direct imaging of black hole shadow halos has firmly confirmed the existence of supermassive black holes (SMBHs), with millions of solar masses, residing at the centers of the Milky Way and M87 galaxies. These groundbreaking discoveries represent a monumental success of Einstein's theory of general relativity and have revealed the hidden "monsters" lurking a

Hao Wu, Yuan Gao, Chang Liu, Fan Xu

Accurately predicting the long-term evolution of turbulence is crucial for advancing scientific understanding and optimizing engineering applications. However, existing deep learning methods face significant bottlenecks in long-term autoregressive prediction, which exhibit excessive smoothing and fail to accurately track complex fluid dynamics. Our extensive

Jingyang Wu, Rohitash Chandra

The rising number of extreme climate events in the past decades has motivated the need for a thorough consideration of tropical cyclone genesis and intensity, given the sea-surface temperature (SST). In this paper, we present an analysis of the relationship between the increasing global SST with cyclone genesis using linear regression machine learning models

Ke-Han Lu, Chun-Yi Kuan, Hung-yi Lee

We introduce Speech-IFeval, an evaluation framework designed to assess instruction-following capabilities and quantify catastrophic forgetting in speech-aware language models (SLMs). Recent SLMs integrate speech perception with large language models (LLMs), often degrading textual capabilities due to speech-centric training. Existing benchmarks conflate spee

Hanfei Zhou, Lei Shi

Physics-informed neural networks (PINNs) provide a mesh-free approach to solving high-dimensional PDEs on complex geometries, but their theoretical foundations on manifolds remain limited. Moreover, conventional PINN analyses typically rely on solution smoothness, while PINNs may perform poorly for low-regularity solutions arising from nonlinear hyperbolic e

Ahmad Moussavi, Peter Danchev, Arash Javan, Omid Hasanzadeh

We give a comprehensive study of the so-called \textit{semi-tripotent rings} obtaining their new and non-trivial characterization as well as a complete description in terms of sums and products of some special elements. Particularly, we explore in-depth when a group ring is semi-tripotent. Our results somewhat supply those established by Ko$\c{s}$an et al. i

Teng Liu, Andreas Morr, Sebastian Bathiany, Lana L. Blaschke

The resilience, or stability, of major Earth system components is increasingly threatened by anthropogenic pressures, demanding reliable early warning signals for abrupt and irreversible regime shifts. Widely used data-driven resilience indicators based on variance and autocorrelation detect `critical slowing down', a signature of decreasing stability. Howev

Yuanyuan Xing, Zihao Zhang

This paper concerns the well-posedness of subsonic Euler-Poisson flows in a convergent nozzle. Due to the geometry of the nozzle, we first introduce a coordinate transformation to prove the existence of radially symmetric subsonic solutions to the steady Euler-Poisson system. We then investigate the structural stability of these background subsonic flows und

Xikai Yang, Juzheng Miao, Yuchen Yuan, Jiaze Wang

Medical large vision-language models (LVLMs) have demonstrated promising performance across various single-image question answering (QA) benchmarks, yet their capability in processing multi-image clinical scenarios remains underexplored. Unlike single image based tasks, medical tasks involving multiple images often demand sophisticated visual understanding c

Zhengkang Guo, Wenhao Liu, Mingchen Xie, Jingwen Xu

Large language models (LLMs) are increasingly expected to tackle complex tasks, driven by their expanding applications and users&#39; growing proficiency in crafting sophisticated prompts. However, as the number of explicitly stated requirements increases (particularly more than 10 constraints), LLMs often struggle to accurately follow such complex instructi

Eoin Ó Colgáin, Saeed Pourojaghi, M. M. Sheikh-Jabbari

We continue scientific scrutiny of the DESI dynamical dark energy (DE) claim by explicitly demonstrating that the result depends on the analysis pipeline. Concretely, we define a likelihood that converts the $w_0 w_a$CDM model back into the (flat) $\Lambda$CDM model, which we fit to DESI constraints on the $\Lambda$CDM model from DR1 Full-Shape (FS) modellin

Tianchi Xie, Minzhi Lin, Mengchen Liu, Yilin Ye

Understanding infographic charts with design-driven visual elements (e.g., pictograms, icons) requires both visual recognition and reasoning, posing challenges for multimodal large language models (MLLMs). However, existing visual-question answering benchmarks fall short in evaluating these capabilities of MLLMs due to the lack of paired plain charts and vis

Dongxu Chang, Qingqing Peng, Guanghui Wang, Guiying Yan

In this study, a scheduling policy of layered decoding for quasi-cycle (QC) low-density parity-check (LDPC) codes with high throughput and good performance is designed. The influence of scheduling on the delay of the decoder's hardware implementation and on the decoding performance are considered simultaneously. Specifically, we analyze the idle time require

Weifeng Kong, Zhiying Tan

Staircase is one of the most common structures in artificial scenes. However, it is difficult for humanoid robots and people with lower limb disabilities or visual impairment to cross the scene without the help of sensors and intelligent algorithms. Staircase scene perception technology is a prerequisite for recognition and localization. This technology is o

Mushtari Sadia, Zhenning Yang, Yunming Xiao, Ang Chen

Relational databases are central to modern data management, yet most data exists in unstructured forms like text documents. To bridge this gap, we leverage large language models (LLMs) to automatically synthesize a relational database by generating its schema and populating its tables from raw text. We introduce SQUiD, a novel neurosymbolic framework that de

Siqi Huang, Yanchen Xu, Hongyuan Zhang, Xuelong Li

Although graph contrastive learning (GCL) has been widely investigated, it is still a challenge to generate effective and stable graph augmentations. Existing methods often apply heuristic augmentation like random edge dropping, which may disrupt important graph structures and result in unstable GCL performance. In this paper, we propose Positive-incentive N

Huda Alghoraibi, Nuha Alqurashi, Sarah Alotaibi, Renad Alkhudaydi

Monkeypox is a viral disease characterized by distinctive skin lesions and has been reported in many countries. The recent global outbreak has emphasized the urgent need for scalable, accessible, and accurate diagnostic solutions to support public health responses. In this study, we developed ITMAINN, an intelligent, AI-driven healthcare system specifically