May 2025 arXiv papers — page 34

Mark Fortune, Neale P. Gibson, Hannah Diamond-Lowe, João M. Mendonça

Time-series photometry at mid-infrared wavelengths is becoming a common technique to search for atmospheres around rocky exoplanets. This method constrains the brightness temperature of the planet to determine whether heat redistribution is taking place - indicative of an atmosphere - or whether the heat is reradiated from a low albedo bare rock. By observin

V. V. Val'kov, A. O. Zlotnikov, A. Gamov, N. A. Fedorova

It is shown that the experimentally detected features in the low-temperature behavior of the magnetization in an external magnetic field perpendicular to the layers of manganese ions of the topological antiferromagnet MnBi$_2$Te$_4$ are due to quantum effects induced by the off-diagonal nature of the trigonal component of the crystal field. In this case, the

Xuchen Ma, Jianxiang Yu, Wenming Shao, Bo Pang

Social media platforms have experienced a significant rise in toxic content, including abusive language and discriminatory remarks, presenting growing challenges for content moderation. Some users evade censorship by deliberately disguising toxic words through homophonic cloak, which necessitates the task of unveiling cloaked toxicity. Existing methods are m

Eddy Divin Kenvo Songwa, Shaday Jesus Nobosse Nguemeta, Hodaya Gabber, Renana Aharonof

Decoupling the global Berry-curvature contribution to the anomalous Hall conductivity (AHC) from local domain- and texture-related contributions in bulk ferromagnetic Weyl semimetals is difficult in standard measurements. We address this in a $\sim$670$\mu$m-thick Co$_3$Sn$_2$S$_2$ single crystal using a contact architecture that promotes depth-distributed c

Philipp Ertz, Petra Friederichs

The aim of this study is to provide a probabilistic gust analysis for the region of Germany that is calibrated with station observations and with an interpolation to unobserved locations. To this end, we develop a spatial Bayesian hierarchical model (BHM) for the post-processing of surface maximum wind gusts from the COSMO-REA6 reanalysis. Our approach uses

Márton Hajdu, Robin Coutelier, Laura Kovács, Andrei Voronkov

The superposition calculus for reasoning in first-order logic with equality relies on simplification orderings on terms. Modern saturation provers use the Knuth-Bendix order (KBO) and the lexicographic path order (LPO) for discovering redundant clauses and inferences. Implementing term orderings is however challenging. While KBO comparisons can be performed

Tuyen Trung Truong

In this paper, we introduce some new iterative optimisation algorithms on Riemannian manifolds and Hilbert spaces which have good global convergence guarantees to local minima. More precisely, these algorithms have the following properties: If $\{x_n\}$ is a sequence constructed by one such algorithm then: - Finding critical points: Any cluster point of $\{x

Yudi Zhang, Weilin Zhao, Xu Han, Tiejun Zhao

Speculative decoding and quantization effectively accelerate memory-bound inference of large language models. Speculative decoding mitigates the memory bandwidth bottleneck by verifying multiple tokens within a single forward pass, which increases computational effort. Quantization achieves this optimization by compressing weights and activations into lower

Vincent Astier, Thomas Unger

We introduced positive cones in an earlier paper as a notion of ordering on central simple algebras with involution that corresponds to signatures of hermitian forms. In the current paper we describe signatures of hermitian forms directly out of positive cones, and also use this approach to rectify a problem that affected some results in the previously menti

Lei Liu, Yanmei Xiao, Tao Guo

Heavy flavor hadrons, especially doubly heavy baryons and doubly heavy tetraquarks, have always received extensive attention in theoretical and experimental research. Given the separation of quark masses $m_Q \gg m_q$ ($Q = c, b$ and $q = u, d, s$), this type of heavy flavor hadrons can be well regarded as hydrogen-like structures in the strong interaction.

Vihang Pancholi, Jainit Bafna, Tejas Anvekar, Manish Shrivastava

Evaluating tables qualitatively and quantitatively poses a significant challenge, as standard metrics often overlook subtle structural and content-level discrepancies. To address this, we propose a rubric-based evaluation framework that integrates multi-level structural descriptors with fine-grained contextual signals, enabling more precise and consistent ta

Hayate Kojima, Keigo Takanami, Junya Hara, Yukihiro Bandoh

We propose a denoising method for multimodal graph signals by an alternating minimization scheme that sequentially solves signal restoration and graph learning problems. Many complex-structured data, i.e., those on sensor networks, can capture multiple modalities at each measurement point, referred to as modalities. They are also assumed to have an underlyin

Georgios Amanatidis, Alexandros Lolos, Evangelos Markakis, Victor Turmel

We study an online fair division setting, where goods arrive one at a time and there is a fixed set of $n$ agents, each of whom has an additive valuation function over the goods. Once a good appears, the value each agent has for it is revealed and it must be allocated immediately and irrevocably to one of the agents. It is known that without any assumptions

Xiang Huang, Ting-En Lin, Feiteng Fang, Yuchuan Wu

Instruction following (IF) is a critical capability for large language models (LLMs). However, handling complex instructions with multiple constraints remains challenging. Previous methods typically select preference pairs based on the number of constraints they satisfy, introducing noise where chosen examples may fail to follow some constraints and rejected

Fatimah Rita Ahmadi

Unitary Ribbon Fusion Categories (URFC) formalize anyonic theories. It has been widely assumed that the same category formalizes a topological quantum computing model. However, in previous work, we addressed and resolved this confusion and demonstrated while the former could be any fusion category, the latter is always a subcategory of Hilb. In this paper, w

Chi Lu, Yiyang Ni, Zhe Wang, Xiaoli Shi

Decision Transformer (DT) has recently demonstrated strong generalizability in dynamic resource allocation within unmanned aerial vehicle (UAV) networks, compared to conventional deep reinforcement learning (DRL). However, its performance is hindered due to zero-padding for varying state dimensions, inability to manage long-term energy constraint, and challe

Moritz René Schäfer, Nico Segreto, Fabian Zills, Christian Holm

We introduce Atomistic learned potentials in JAX (apax), a flexible and efficient open source software package for training and inference of machine-learned interatomic potentials. Built on the JAX framework, apax supports GPU acceleration and implements flexible model abstractions for fast development. With features such as kernel-based data selection, well

Weilun Feng, Chuanguang Yang, Haotong Qin, Xiangqi Li

Diffusion transformers (DiT) have demonstrated exceptional performance in video generation. However, their large number of parameters and high computational complexity limit their deployment on edge devices. Quantization can reduce storage requirements and accelerate inference by lowering the bit-width of model parameters. Yet, existing quantization methods

Kazuki Wataya, Takehisa Hasegawa

We investigate bond percolation on the Song-Havlin-Makse (SHM) network, a scale-free tree with a tunable degree exponent and dimensionality. Using a generating function approach, we analytically derive the average size and the fractal exponent of the root cluster for deterministic cases. Our analysis reveals that bond percolation on the SHM network remains i

Bocheng Li, Zhujin Gao, Linli Xu

Diffusion models have emerged as a promising approach for text generation, with recent works falling into two main categories: discrete and continuous diffusion models. Discrete diffusion models apply token corruption independently using categorical distributions, allowing for different diffusion progress across tokens but lacking fine-grained control. Conti

Yi Liu, Hongji Zhang, Yunhao Zhou, Zhengyuan Shi

The integration of large language models (LLMs) into electronic design automation (EDA) has significantly advanced the field, offering transformative benefits, particularly in register transfer level (RTL) code generation and understanding. While previous studies have demonstrated the efficacy of fine-tuning LLMs for these generation-based tasks, embedding-b

Marcello Baldo

The spontaneous decay of an excited atom by photon emission is one of the most common and elementary physical process present in nature and in laboratories. The decay is random in time with constant probability density, as it can be inferred by the exponential law observed experimentally. Despite the simplicity of the process, in Quantum Mechanics the decay

Vladimir Dzhunushaliev, Vladimir Folomeev

The physical effect of condensation of a classical spinor field at the event horizon is under consideration. The corresponding solution is sought for the set of the Einstein-Dirac equations. It is shown that in this case there arises a black hole with a $\delta$-like classical spinor field concentrated at the event horizon.

Hongyu Jin, Panos Papadimitratos

Paramount to vehicle safety, broadcasted Cooperative Awareness Messages (CAMs) and Decentralized Environmental Notification Messages (DENMs) are pseudonymously authenticated for security and privacy protection, with each node needing to have all incoming messages validated within an expiration deadline. This creates an asymmetry that can be easily exploited

A. M. Kutkin, R. Morganti, T. A. Oosterloo, E. A. K. Adams

We present two new radio continuum images obtained with Apertif at 1.4 GHz. The images, produced with a direction-dependent calibration pipeline, cover 136 square degrees of the Lockman Hole and 24 square degrees of the ELAIS-N fields, with an average resolution of 17x12" and residual noise of 33 uJy/beam. With the improved depth of the images we found in to

Matias Slavov

This paper shall explore the conjunction of eternalism and Everettian quantum mechanics. It shall be argued that there is a strong analogy between these two views. In case there is an indefinite number of worlds and observers that are all equally real, there should be an indefinite number of local times which are all also equally real. Whereas Everettianism,

Jiawen Yu, Hairuo Liu, Qiaojun Yu, Jieji Ren

Vision-Language-Action (VLA) models have advanced general-purpose robotic manipulation by leveraging pretrained visual and linguistic representations. However, they struggle with contact-rich tasks that require fine-grained control involving force, especially under visual occlusion or dynamic uncertainty. To address these limitations, we propose ForceVLA, a

Rustem Takhanov

In the past decade gradient-based deep learning has revolutionized several applications. However, this rapid advancement has highlighted the need for a deeper theoretical understanding of its limitations. Research has shown that, in many practical learning tasks, the information contained in the gradient is so minimal that gradient-based methods require an e

Paramita Mirza, Lucas Weber, Fabian Küch

Recent work shows that post-training datasets for LLMs can be substantially downsampled without noticeably deteriorating performance. However, data selection often incurs high computational costs or is limited to narrow domains. In this paper, we demonstrate that data selection can be both -- efficient and universal -- by using a multi-step pipeline in which

Shuaiyi Li, Zhisong Zhang, Yang Deng, Chenlong Deng

Although existing model editing methods perform well in recalling exact edit facts, they often struggle in complex scenarios that require deeper semantic understanding rather than mere knowledge regurgitation. Leveraging the strong contextual reasoning abilities of large language models (LLMs), in-context learning (ICL) becomes a promising editing method by

Sascha Rechenberger, Thom Frühwirth

Constraint Handling Rules (CHR) is a rule-based programming language which is typically embedded into a general-purpose language. There exists a plethora of implementations of CHR for numerous host languages. However, the existing implementations often reinvent the way to embed CHR, which impedes maintenance and weakens assertions of correctness. To formaliz

Chao Tian, Chao Yang, Guoqing Zhu, Qiang Wang

RGB-Thermal (RGB-T) object detection utilizes thermal infrared (TIR) images to complement RGB data, improving robustness in challenging conditions. Traditional RGB-T detectors assume balanced training data, where both modalities contribute equally. However, in real-world scenarios, modality degradation-due to environmental factors or technical issues-can lea

Xiaokai Chen, Xiao Lin, Changcheng Li, Peng Jiang

In online video platforms, accurate watch time prediction has become a fundamental and challenging problem in video recommendation. Previous research has revealed that the accuracy of watch time prediction highly depends on both the transformation of watch-time labels and the decomposition of the estimation process. TPM (Tree based Progressive Regression Mod

Dominik Fuchsgruber, Tom Wollschläger, Johannes Bordne, Stephan Günnemann

While uncertainty estimation for graphs recently gained traction, most methods rely on homophily and deteriorate in heterophilic settings. We address this by analyzing message passing neural networks from an information-theoretic perspective and developing a suitable analog to data processing inequality to quantify information throughout the model's layers.

Claude Formanek, Omayma Mahjoub, Louay Ben Nessir, Sasha Abramowitz

A key challenge in offline multi-agent reinforcement learning (MARL) is achieving effective many-agent multi-step coordination in complex environments. In this work, we propose Oryx, a novel algorithm for offline cooperative MARL to directly address this challenge. Oryx adapts the recently proposed retention-based architecture Sable and combines it with a se

Runze Xia, Shuo Feng, Renzhi Wang, Congchi Yin

Brain-to-Image reconstruction aims to recover visual stimuli perceived by humans from brain activity. However, the reconstructed visual stimuli often missing details and semantic inconsistencies, which may be attributed to insufficient semantic information. To address this issue, we propose an approach named Fine-grained Brain-to-Image reconstruction (FgB2I)

Jan Danek, Zdenek Becvar, Adam Janes

We focus on computation offloading of applications based on convolutional neural network (CNN) from moving devices, such as mobile robots or autonomous vehicles, to MultiAccess Edge Computing (MEC) servers via a mobile network. In order to reduce overall CNN inference time, we design and implement CNN with early exits and splits, allowing a flexible partial

Valentine Bernasconi, Gustavo Marfia

Recent breakthroughs in generative AI have opened the door to new research perspectives in the domain of art and cultural heritage, where a large number of artifacts have been digitized. There is a need for innovation to ease the access and highlight the content of digital collections. Such innovations develop into creative explorations of the digital image

Gangwei Jiang, Yahui Liu, Zhaoyi Li, Qi Wang

Recent advances in reasoning with large language models (LLMs) have popularized Long Chain-of-Thought (LCoT), a strategy that encourages deliberate and step-by-step reasoning before producing a final answer. While LCoTs have enabled expert-level performance in complex tasks, how the internal structures of their reasoning chains drive, or even predict, the co

Florian Andreas Marwitz, Tanya Braun, Ralf Möller, Marcel Gehrke

When allowing concurrent actions in Markov Decision Processes, whose state and action spaces grow exponentially in the number of objects, computing a policy becomes highly inefficient, as it requires enumerating the joint of the two spaces. For the case of indistinguishable objects, we present a first-order representation to tackle the exponential blow-up in

Guangfu Hao, Haojie Wen, Liangxuan Guo, Yang Chen

Flexible tool selection reflects a complex cognitive ability that distinguishes humans from other species, yet computational models that capture this ability remain underdeveloped. We developed a framework using low-dimensional attribute representations to bridge visual tool perception and linguistic task understanding. We constructed a comprehensive dataset

Foivos Evangelopoulos-Ntemiris, Mark Veraar

In this paper, we investigate discrete regularity estimates for a broad class of temporal numerical schemes for parabolic stochastic evolution equations. We provide a characterization of discrete stochastic maximal $\ell^p$-regularity in terms of its continuous counterpart, thereby establishing a unified framework that yields numerous new discrete regularity

Hisham Sati, Urs Schreiber

Fractional quantum Hall systems (FQH), due to their experimentally observed anyonic topological order, are a main contender for future hardware-implementation of error-protected quantum registers ("topological qbits") subject to error-protected quantum operations ("topological quantum gates"), both plausibly necessary for future quantum computing at useful s

Fengyun Wang, Sicheng Yu, Jiawei Wu, Jinhui Tang

Large vision-language models (LVLMs) have significantly advanced numerous fields. In this work, we explore how to harness their potential to address 3D scene understanding tasks, using 3D question answering (3D-QA) as a representative example. Due to the limited training data in 3D, we do not train LVLMs but infer in a zero-shot manner. Specifically, we samp

Keita Hidaka, Dina Abdelhadi, Ruediger Urbanke

Good quantum error-correcting codes that fulfill practical considerations, such as simple encoding circuits and efficient decoders, are essential for functional quantum information processing systems. Quantum polar codes satisfy some of these requirements but lack certain critical features, thereby hindering their widespread use. Existing constructions eithe

Guanwen Feng, Zhiyuan Ma, Yunan Li, Jiahao Yang

Recent advances in audio-driven talking head generation have achieved impressive results in lip synchronization and emotional expression. However, they largely overlook the crucial task of facial attribute editing. This capability is indispensable for achieving deep personalization and expanding the range of practical applications, including user-tailored di

BESIII Collaboration, M. Ablikim, M. N. Achasov, P. Adlarson

A search for a dark baryon is performed for the first time in the two-body decay $\Xi^-\rightarrow\pi^-+{\rm invisible}$ using $(10.087\pm0.044)\times10^{9}$ $J/\psi$ events collected at a center-of-mass energy of $\sqrt{s}=3.097\,\mbox{GeV}$ with the BESIII detector at the BEPCII collider. No significant signal is observed, and the 90% (95%) confidence leve

Qingchuan Ma, Yuhang Wu, Xiawu Zheng, Rongrong Ji

In this paper, we aim to establish a simple, effective, and theoretically grounded benchmark for rigorously probing abstract reasoning in Large Language Models (LLMs). To achieve this, we first develop a mathematic framework that defines abstract reasoning as the ability to: (i) extract essential patterns independent of surface representations, and (ii) appl

Chaeeun Kim, Jinu Lee, Wonseok Hwang

Legal Case Retrieval (LCR), which retrieves relevant cases from a query case, is a fundamental task for legal professionals in research and decision-making. However, existing studies on LCR face two major limitations. First, they are evaluated on relatively small-scale retrieval corpora (e.g., 100-55K cases) and use a narrow range of criminal query types, wh

Okwudili D. Egbo, David A. H. Buckley, Paul J. Groot, Francesco Cavallaro

We report on optically selected stellar candidates of SARAO MeerKAT 1.3 GHz radio continuum survey sources of the Galactic plane. Stellar counterparts to radio sources are selected by cross-matching the MeerKAT source positions with \textit{Gaia} DR3, using two approaches. The first approach evaluated the probability of chance alignments between the radio su

Wenhao Ye, Tiansheng Zheng, Yue Qi, Wenhua Zhao

The intangible cultural heritage (ICH) of China, a cultural asset transmitted across generations by various ethnic groups, serves as a significant testament to the evolution of human civilization and holds irreplaceable value for the preservation of historical lineage and the enhancement of cultural self-confidence. However, the rapid pace of modernization p

S. A. Avdonin, A. Choque Rivero, G. Leugering, V. S. Mikhaylov

In this article the authors continue the discussion in \cite{ALM} about inverse problems for second order elliptic and hyperbolic equations on metric trees from boundary measurements. In the present paper we prove the identifiability of varying densities of a planar tree-like network of strings along with the complete information on the graph, i.e. the lengt

Marc Feger, Katarina Boland, Stefan Dietze

Identifying arguments is a necessary prerequisite for various tasks in automated discourse analysis, particularly within contexts such as political debates, online discussions, and scientific reasoning. In addition to theoretical advances in understanding the constitution of arguments, a significant body of research has emerged around practical argument mini

Zheng-Yi Lu

In this paper, we investigate the relationship between frame bounds and spectral gaps. By introducing the notion of \emph{essential minimum(maximal) spectral gap}, we provide a local characterization of Landau's theorem \cite{Lan67}. As an application, we resolve the spectrality additive measures of Lebesgue type, conclusively answering an open question on t

Yuichiro Hoshino, Hideyuki Tachibana, Muneyoshi Inahara, Hiroto Takegawa

Hybrid models combining Transformers and State Space Models (SSMs) are promising for balancing performance and efficiency. However, optimizing these hybrid models, particularly by addressing the potential redundancy inherent within the Transformer components, remains a significant challenge. In this paper, we propose RAD (Redundancy-Aware Distillation), a no

Seong Jun Park, M. Y. Choi

In general, the rates of infection and removal (whether through recovery or death) are nonlinear functions of the number of infected and susceptible individuals. One of the simplest models for the spread of infectious diseases is the SIR model, which categorizes individuals as susceptible, infectious, recovered or deceased. In this model, the infection rate,

Tiantian Feng, Thanathai Lertpetchpun, Dani Byrd, Shrikanth Narayanan

Speech emotion recognition (SER), particularly for naturally expressed emotions, remains a challenging computational task. Key challenges include the inherent subjectivity in emotion annotation and the imbalanced distribution of emotion labels in datasets. This paper introduces the \texttt{SAILER} system developed for participation in the INTERSPEECH 2025 Em

Daniel Mejías, Inhar Yeregui, Ángel Martín, Roberto Viola

The proliferation of Extended Reality (XR) applications, requiring high-quality, low-latency media streaming, has driven the demand for efficient remote rendering solutions. This paper focuses on holographic conferencing in virtual environments and their required uplink and downlink media transmission capabilities. By examining Media over QUIC (MoQ), Real-ti

Zhuoyang Wu, Xinze Li, Zhenghao Liu, Yukun Yan

Large Language Models (LLMs) have exhibited strong reasoning capabilities and achieved remarkable performance in mathematical problem-solving tasks. Recently, distilling reasoning ability from long-form Chains-of-Thought (CoTs) has emerged as a promising approach for enhancing Small Language Models (SLMs). Existing studies typically treat SLMs as student mod

Haidong Xin, Zhenghao Liu, Sen Mei, Yukun Yan

User-item interaction histories are pivotal for sequential recommendation systems but often include noise, such as unintended clicks or actions that fail to reflect genuine user preferences. To address this, we propose Learned Item Shortcuts for Sequential Recommendation (LISRec), a novel framework that explicitly captures stable preferences by extracting pe

Jinhong Ni, Chang-Bin Zhang, Qiang Zhang, Jing Zhang

Recent prosperity of text-to-image diffusion models, e.g. Stable Diffusion, has stimulated research to adapt them to 360-degree panorama generation. Prior work has demonstrated the feasibility of using conventional low-rank adaptation techniques on pre-trained diffusion models to generate panoramic images. However, the substantial domain gap between perspect

Alejandro D. Mousist

This work addresses mechanical defocus in Earth observation images from the IMAGIN-e mission aboard the ISS, proposing a blind deblurring approach adapted to space-based edge computing constraints. Leveraging Sentinel-2 data, our method estimates the defocus kernel and trains a restoration model within a GAN framework, effectively operating without reference

Lara Bossinger, Jacinta Torres

We present n-1 different embeddings of string polytopes of type A. We characterize their compatibility with the crystal structure on the string polytopes, and formulate a conjecture describing how to obtain n-1 different atomic decompositions of the crystal with highest weight a multiple of the highest root.

Yifan Chang, Yukang Feng, Jianwen Sun, Jiaxin Ai

Recent years have seen rapid advances in AI-driven image generation. Early diffusion models emphasized perceptual quality, while newer multimodal models like GPT-4o-image integrate high-level reasoning, improving semantic understanding and structural composition. Scientific illustration generation exemplifies this evolution: unlike general image synthesis, i

Melrose Tia, Jezreel Sophia Lanuzo, Lei Rigi Baltazar, Marie Joy Lopez-Relente

Traditional sentiment analysis relies on surface-level linguistic patterns and retrospective data, limiting its ability to capture the psychological and contextual drivers of human sentiment. These limitations constrain its effectiveness in applications that require predictive insight, such as policy testing, narrative framing, and behavioral forecasting. We

Si Zhang, Paul Mingzheng Tang, Hoong Chuin Lau

Nurse staffing and scheduling are persistent challenges in healthcare due to demand fluctuations and individual nurse preferences. This study introduces the concept of bounded flexibility, balancing nurse satisfaction with strict rostering rules, particularly a real-world time regularity policy from a major hospital in Singapore. We model the problem as a mu

Inhar Yeregui, Daniel Mejías, Mikel Zorrilla, Roberto Viola

As immersive eXtended Reality (XR) applications demand substantial network resources, understanding their interaction with 5G networks becomes crucial to improve them. This paper investigates the role of 5G physical-layer monitoring to manage and enhance the remote rendering of XR content dynamically. By observing network metrics directly from the physical l

Duy-Minh Dang, Chang Chen

We investigate multi-period mean-risk portfolio optimization for long-horizon Defined Contribution plans, focusing on buffered Probability of Exceedance (bPoE), a more intuitive, dollar-based alternative to Conditional Value-at-Risk (CVaR). We formulate both pre-commitment and time-consistent Mean-bPoE and Mean-CVaR portfolio optimization problems under real

Runyu Wang, Peng Ping, Zhengyu Guo, Xiaoye Zhang

Fine-tuning adapts pretrained models for specific tasks but poses the risk of catastrophic forgetting (CF), where critical knowledge from pretraining is overwritten. To address the issue of CF in a general-purpose framework, we propose Low-damage Knowledge Implanting (LoKI), a parameter-efficient fine-tuning (PEFT) technique that utilizes recent mechanistic

Nabamita Banerjee, Jitesh Singh

We study massive self-interacting vector field theories with mass added ``by hand". We show that the massless limit of the quartic self-interacting vector field theory is not smooth. Pathological behavior of the theory is not limited only at the classical level, even at the quantum level unitarity is violated at the two-loop. Using the Vainshtein mechanism,

Alan Ramponi, Marco Rovera, Robert Moro, Sara Tonelli

Retrieval of previously fact-checked claims is a well-established task, whose automation can assist professional fact-checkers in the initial steps of information verification. Previous works have mostly tackled the task monolingually, i.e., having both the input and the retrieved claims in the same language. However, especially for languages with a limited

Asunción Fuente, Gisela Esplugues, Pablo Rivière-Marichalar, David Navarro-Almaida

Sulfur is essential for life, but its abundance and distribution in the interstellar medium remain uncertain, with over 90% of sulfur undetected in cold molecular clouds. Sulfur allotropes (S$_{\rm n}$) have been proposed as possible reservoirs, but the only detected interstellar molecule with a disulfide bond is S$_2$H in the Horsehead Nebula, making the es

Jintao Zhang, Zirui Liu, Mingyue Cheng, Shilong Zhang

Intraoperative hypotension (IOH) frequently occurs under general anesthesia and is strongly linked to adverse outcomes such as myocardial injury and increased mortality. Despite its significance, IOH prediction is hindered by event sparsity and the challenge of integrating static and dynamic data across diverse patients. In this paper, we propose \textbf{IOH

E Fan, Kang Hu, Zhuowen Wu, Jiangyang Ge

Computational Fluid Dynamics (CFD) is critical for scientific advancement but is hindered by operational complexity and high expertise barriers. This paper introduces ChatCFD, a Large Language Model (LLM)-driven multi-agent system designed for end-to-end CFD automation using OpenFOAM. Powered by DeepSeek-R1/V3, ChatCFD integrates structured domain knowledge

S. A. Avdonin, G. Leugering, V. S. Mikhaylov

We consider the in-plane motion of elastic strings on tree-like network, observed from the 'leaves'. We investigate the inverse problem of recovering not only the physical properties i.e. the 'optical lengths' of each string, but also the topology of the tree which is represented by the edge degrees and the angles between branching edges. To this end use the

MaryBeth Defrance, Guillaume Bied, Maarten Buyl, Jefrey Lijffijt

Over the past 15 years, hundreds of bias mitigation methods have been proposed in the pursuit of fairness in machine learning (ML). However, algorithmic biases are domain-, task-, and model-specific, leading to a `portability trap': bias mitigation solutions in one context may not be appropriate in another. Thus, a myriad of design choices have to be made wh

Zhiyuan Li, Yi Chang, Yuan Wu

Large reasoning models (LRMs) have achieved impressive performance in complex tasks, often outperforming conventional large language models (LLMs). However, the prevalent issue of overthinking severely limits their computational efficiency. Overthinking occurs when models generate excessive and redundant tokens that contribute little to accurate outcomes, es

Guangfu Hao, Frederic Alexandre, Shan Yu

Cognitive flexibility has been extensively studied in human cognition but remains relatively unexplored in the context of Visual Large Language Models (VLLMs). This study assesses the cognitive flexibility of state-of-the-art VLLMs (GPT-4o, Gemini-1.5 Pro, and Claude-3.5 Sonnet) using the Wisconsin Card Sorting Test (WCST), a classic measure of set-shifting

Woonho Ko, Jin Bok Park, Il Yong Chun

Existing long-term video prediction methods often rely on an autoregressive video prediction mechanism. However, this approach suffers from error propagation, particularly in distant future frames. To address this limitation, this paper proposes the first AutoRegression-Free (ARFree) video prediction framework using diffusion models. Different from an autore

Yue Cui, Liuyi Yao, Shuchang Tao, Weijie Shi

Large language models (LLMs) have significantly advanced natural language processing, particularly through the integration of external tools and APIs. However, their effectiveness is frequently hampered by parameter mis-filling during tool calling. In this paper, we propose the Hierarchical Tool Error Checklist (HiTEC) framework to systematically diagnose an

Linglin Jing, Yuting Gao, Zhigang Wang, Wang Lan

Recent advancements have shown that the Mixture of Experts (MoE) approach significantly enhances the capacity of large language models (LLMs) and improves performance on downstream tasks. Building on these promising results, multi-modal large language models (MLLMs) have increasingly adopted MoE techniques. However, existing multi-modal MoE tuning methods ty

Inmaculada Gayte Delgado, Irene Marín Gayte

This work obtains a fixed-point equation for the solution of linear parabolic partial differential problems based on solutions to heat problems. This is a pointwise equality, so we have required non-standard techniques that involve the study of the sign of certain solutions to linear parabolic problems. This fixed-point equation implies regularity properties

Paul Krzakala, Gabriel Melo, Charlotte Laclau, Florence d'Alché-Buc

Although graph-based learning has attracted a lot of attention, graph representation learning is still a challenging task whose resolution may impact key application fields such as chemistry or biology. To this end, we introduce GRALE, a novel graph autoencoder that encodes and decodes graphs of varying sizes into a shared embedding space. GRALE is trained u

Shuhai Zhang, Zeng You, Yaofo Chen, Zhiquan Wen

Transformer-based large language models (LLMs) excel in natural language processing tasks by capturing long-range dependencies through self-attention mechanisms. However, long-context modeling faces significant computational inefficiencies due to \textit{redundant} attention computations: while attention weights are often \textit{sparse}, all tokens consume

Junqi Zhao, Jinzheng Zhao, Haohe Liu, Yun Chen

Diffusion models have significantly improved the quality and diversity of audio generation but are hindered by slow inference speed. Rectified flow enhances inference speed by learning straight-line ordinary differential equation (ODE) paths. However, this approach requires training a flow-matching model from scratch and tends to perform suboptimally, or eve

Hang Chen, Maoyuan Ye, Peng Yang, Haibin He

Power transmission corridor hazard segmentation (PTCHS) aims to separate transmission equipment and surrounding hazards from complex background, conveying great significance to maintaining electric power transmission safety. Recently, the Segment Anything Model (SAM) has emerged as a foundational vision model and pushed the boundaries of segmentation tasks.

Davide Corsi, Kaushik Mallik, Andoni Rodriguez, Cesar Sanchez

Shielding has emerged as a promising approach for ensuring safety of AI-controlled autonomous systems. The algorithmic goal is to compute a shield, which is a runtime safety enforcement tool that needs to monitor and intervene the AI controller's actions if safety could be compromised otherwise. Traditional shields are designed statically for a specific

Martin J. Gander, Liu-Di Lu, Tingting Wu

We present here nonoverlapping optimized Schwarz methods applied to heat transfer problems with heterogeneous diffusion coefficients. After a Laplace transform in time, we derive the error equation and obtain the convergence factor. The optimal transmission operators are nonlocal, and thus inconvenient to use in practice. We introduce three versions of local

Paul Disberg, Ilya Mandel

Neutron stars (NSs) are thought to receive natal kicks at their formation in supernovae. In order to investigate the magnitude of these kicks, we analyze the proper motions and distance estimates -- either through parallax or dispersion measures -- of young isolated pulsars and infer their three-dimensional velocities relative to their local standard of rest

Zhiyu Li, Shichao Song, Hanyu Wang, Simin Niu

Large Language Models (LLMs) have emerged as foundational infrastructure in the pursuit of Artificial General Intelligence (AGI). Despite their remarkable capabilities in language perception and generation, current LLMs fundamentally lack a unified and structured architecture for handling memory. They primarily rely on parametric memory (knowledge encoded in

Qian Chen, Benoît Collins, Omar Fawzi

We study the problem of testing $k$-block-positivity via symmetric $N$-extendibility by taking the tensor product with a $k$-dimensional maximally entangled state. We exploit the unitary symmetry of the maximally entangled state to reduce the size of the corresponding semidefinite programs (SDP). For example, for $k=2$, the SDP is reduced from one block of s

Wenwen Qiang, Ziyin Gu, Lingyu Si, Jiangmeng Li

In this paper, we addressed the limitation of relying solely on distribution alignment and source-domain empirical risk minimization in Unsupervised Domain Adaptation (UDA). Our information-theoretic analysis showed that this standard adversarial-based framework neglects the discriminability of target-domain features, leading to suboptimal performance. To br

Akshaya Rajesh, Sumbul Khan

The Feynman learning technique is an active learning strategy that helps learners simplify complex information through student-led teaching and discussion. In this paper, we present the development and usability testing of the Feynman Bot, which uses the Feynman technique to assist self-regulated learners who lack peer or instructor support. The Bot embodies

Junhuan Liu, San Jiang, Wei Ge, Wei Huang

The primary contribution of this paper is a challenging benchmark dataset, UAVPairs, and a training pipeline designed for match pair retrieval of large-scale UAV images. First, the UAVPairs dataset, comprising 21,622 high-resolution images across 30 diverse scenes, is constructed; the 3D points and tracks generated by SfM-based 3D reconstruction are employed

Philippe Mounaix

This paper gives a pedestrian presentation of some technical results recently published in mathematical physics with non-trivial implications for laser-plasma interaction. The aim is to get across the main results without going into the details of the calculations, nor offering a specialist's user guide, but by focusing conceptually on how these results modi

Jinheon Baek, Horst Samulowitz, Oktie Hassanzadeh, Dharmashankar Subramanian

Text-to-SQL aims to translate natural language queries into SQL statements, which is practical as it enables anyone to easily retrieve the desired information from databases. Recently, many existing approaches tackle this problem with Large Language Models (LLMs), leveraging their strong capability in understanding user queries and generating corresponding S

Chunyi Peng, Zhipeng Xu, Zhenghao Liu, Yishan Li

Multimodal Retrieval-Augmented Generation (MRAG) has shown promise in mitigating hallucinations in Multimodal Large Language Models (MLLMs) by incorporating external knowledge. However, existing methods typically adhere to rigid retrieval paradigms by mimicking fixed retrieval trajectories and thus fail to fully exploit the knowledge of different retrieval e

Tonghe Zhang, Chao Yu, Sichang Su, Yu Wang

We propose ReinFlow, a simple yet effective online reinforcement learning (RL) framework that fine-tunes a family of flow matching policies for continuous robotic control. Derived from rigorous RL theory, ReinFlow injects learnable noise into a flow policy's deterministic path, converting the flow into a discrete-time Markov Process for exact and straightfor

Santiago Berrezueta-Guzman, Stephan Krusche, Stefan Wagner

The rapid adoption of AI powered coding assistants like ChatGPT and other coding copilots is transforming programming education, raising questions about assessment practices, academic integrity, and skill development. As educators seek alternatives to traditional grading methods susceptible to AI enabled plagiarism, structured peer assessment could be a prom

Valentin Cuzin-Rambaud, Emilien Komlenovic, Alexandre Faure, Bruno Yun

The alignment between humans and machines is a critical challenge in artificial intelligence today. Reinforcement learning, which aims to maximize a reward function, is particularly vulnerable to the risks associated with poorly designed reward functions. Recent advancements has shown that Large Language Models (LLMs) for reward generation can outperform hum