March 2026 arXiv papers — page 82

Sai Koneru, Jian Wu, Sarah Rajtmajer

Extracting hypotheses and their supporting statistical evidence from full-text scientific articles is central to the synthesis of empirical findings, but remains difficult due to document length and the distribution of scientific arguments across sections of the paper. The work studies a sequential full-text extraction setting, where the statement of a prima

Zhiyang Tang, Yiming Zhu, Ruimin Huang, Meng Yang

Due to the limitations of optical lens focal length and detector resolution, distant clustered infrared small targets often appear as mixed spots. The Close Small Object Unmixing (CSOU) task aims to recover the number, sub-pixel positions, and radiant intensities of individual targets from these spots, which is a highly ill-posed inverse problem. Existing me

Rustem Islamov, Roman Machacek, Aurelien Lucchi, Antonio Silveti-Falls

We study the role of batch size in stochastic conditional gradient methods under a $\mu$-Kurdyka-{\L}ojasiewicz ($\mu$-KL) condition. Focusing on momentum-based stochastic conditional gradient algorithms (e.g., Scion), we derive a new analysis that explicitly captures the interaction between stepsize, batch size, and stochastic noise. Our study reveals a reg

Yiwei Wu, Yifan Wu, Yunhao Xiong, Dengwei Zhao

Constructing behavioral-level chiplet models (e.g., SystemC) is crucial for early-stage heterogeneous architecture exploration. Traditional manual modeling is notoriously time-consuming and error-prone. Recently, Large Language Models (LLMs) have demonstrated immense potential in automating hardware code generation. However, existing LLM-assisted design fram

Varshith Madishetty

Decoding speech information from scalp EEG remains difficult due to low SNR and spatial blurring. We present CIPHER (Conformer-based Inference of Phonemes from High-density EEG Representations), a dual-pathway model using (i) ERP features and (ii) broadband DDA coefficients. On OpenNeuro ds006104 (24 participants, two studies with concurrent TMS), binary art

Bina Patel, Jahnvi Mistry, Ayush Bidlan, Parth Bambhaniya

We analyze the causal structure of horizonless compact objects via the light-cone geometry and conformal compactification of the Joshi-Malafarina-Narayan (JMN-1) and Janis-Newman-Winicour (JNW) spacetimes. Penrose diagrams reveal that JMN-1 undergoes a transition from timelike $(0<M_0<2/3)$ to null $(2/3<M_0<4/5)$ singularities, while JNW remains timelike th

Thuy T. Le, Minh-Binh Tran, Loc H. Nguyen

We study an inverse initial-density problem for a nonlinear diffusive coagulation--fragmentation equation with known coagulation and fragmentation kernels. The objective is to recover the unknown initial particle-size distribution on a finite interval from time-dependent boundary observations of the solution and its size derivative. To solve this inverse pro

Alcides Garat

Ever since a new symmetry was found for the imperfect fluid with vorticity the question of the effect of perturbations on the symmetry itself has been raised. This new symmetry arose when realizing that local four-velocity gauge-like transformations would render the left hand side of the Einstein equations invariant. Because the metric tensor would be invari

Ebasa Temesgen, Nathnael Minyelshowa, Lebsework Negash

The use of unmanned aerial vehicles (UAVs) in precision agriculture has seen a huge increase recently. As such, systems that aim to apply various algorithms on the field need a structured framework of abstractions. This paper defines the various tasks of the UAVs in precision agriculture and model them into an architectural framework. The presented architect

Andrey Pustovit

RAG wastes tokens. We propose Knowledge Packs: pre-computed KV caches that deliver the same knowledge at zero token cost. For causal transformers, the KV cache from a forward pass on text F is identical to what a joint pass on F+q would produce - this follows directly from the causal mask. The equivalence is exact but fragile: wrong chat template formatting

Quanyu Tang

For a graph $G$ of order $n$, let $$ \lambda_1(G)\ge \cdots \ge \lambda_n(G) $$ be the eigenvalues of its adjacency matrix. We prove that every graph $G$ on $n\ge 3$ vertices satisfies $$ \lambda_3(G)\le \frac{n}{3}-1, $$ thereby solving a problem of Nikiforov. The bound is best possible whenever $3\mid n$. Our proof is derived from a more general matrix res

Zhuchao Ji, Junyi Xie, Geng-Rui Zhang

Let $f,g\in\overline{\mathbb{Q}}[z]$ be polynomials of degree $d\geq2$ with disconnected Julia sets. We prove that they have the same Lyapunov exponent $\mathcal{L}_f=\mathcal{L}_g$ if and only if either $f$ and $g$ are intertwined, or $f$ and $\overline{g}$ are intertwined. The analogous result for critical heights is also obtained. As an application, we pr

Miryala Sathvika, Rudra Dhar, Karthik Vaidhyanathan

Architecture views are essential for software architecture documentation, yet their manual creation is labor intensive and often leads to outdated artifacts. As systems grow in complexity, the automated generation of views from source code becomes increasingly valuable. Goal: We empirically evaluate the ability of LLMs and agentic approaches to generate arch

Andrei Baroian, Rutger Berger

Reinforcement learning with verifiable rewards (RLVR) plays a crucial role in expanding the capacities of LLM reasoning, but GRPO-style training is dominated by expensive rollouts and wastes compute on unusable prompts. We propose Prompt Replay, an overhead-free online data selection method for GRPO that reuses prompts only (not trajectories), to preserve on

Zifeng Zhu, Jiaming Han, Jiaxiang Zhao, Minnan Luo

While Diffusion Large Language Models (DLLMs) have demonstrated remarkable capabilities in multi-modal generation, performing precise, training-free image editing remains an open challenge. Unlike continuous diffusion models, the discrete tokenization inherent in DLLMs hinders the application of standard noise inversion techniques, often leading to structura

Kwanyoung Kim, Byeongsu Sim

Reinforcement learning from human feedback (RLHF) has proven effective in aligning large language models with human preferences, inspiring the development of reward-centric diffusion reinforcement learning (RDRL) to achieve similar alignment and controllability. While diffusion models can generate high-quality outputs, RDRL remains susceptible to reward hack

Zhiqiang He

This paper investigates the fundamental mechanisms driving plasticity loss in deep reinforcement learning (RL), a critical challenge where neural networks lose their ability to adapt to non-stationary environments. While existing research often relies on descriptive metrics like dormant neurons or effective rank, these summaries fail to explain the underlyin

Edward Phillips, Fredrik K. Gustafsson, Sean Wu, Anshul Thakur

Selective prediction systems can mitigate harms resulting from language model hallucinations by abstaining from answering in high-risk cases. Uncertainty quantification techniques are often employed to identify such cases, but are rarely evaluated in the context of the wider selective prediction policy and its ability to operate at low target error rates. We

Sekhar Ghosh, Vishvesh Kumar, Tapendu Rana

This article investigates the multiplicity of solutions to the Brezis-Nirenberg problem on smooth bounded domains in the hyperbolic space $\mathbb{B}^N$ for $N \ge 4$. Specifically, we study the critical semilinear equation $-\Delta_{\mathbb{B}^N} u = \lambda u + |u|^{2^*-2}u$ under Dirichlet boundary conditions for $\lambda > \frac{N(N-2)}{4}$. Overcoming t

Elif Ceren Gok Yildirim, Murat Onur Yildirim, Joaquin Vanschoren

The continual learning literature has rapidly shifted from traditional class incremental learning (CIL) techniques to foundation model (FM)-based CIL methods without a clear understanding of how these newer approaches compare to strong, lightweight convolutional baselines. This abrupt transition has created a substantial methodological gap, making it difficu

Chen Zhang, Yuxin Cheng, Chenchen Ding, Shuqi Wang

Zeroth-order (ZO) optimization enables memory-efficient training of neural networks by estimating gradients via forward passes only, eliminating the need for backpropagation. However, the stochastic nature of gradient estimation significantly obscures the training dynamics, in contrast to the well-characterized behavior of first-order methods under Neural Ta

B. Lee, B. Kim, P. Adrich, I. Belosevic

CERN's AD/ELENA ``antimatter factory'' - unique worldwide - serves several experiments, all of which use electromagnetic traps to accumulate antiprotons for fundamental science. The GBAR experiment employs a charge-exchange reaction between an antiproton beam and a positronium cloud to produce antihydrogen for gravitational studies. GBAR has also pioneered a

Dengfeng Wang, Shunqin Cai, Yanan Sun

3D point cloud neural networks have significantly enhanced the perceptual capabilities of resource-limited mobile intelligent systems. However, despite the transformative impact, the point cloud algorithm suffers from substantial memory access during data preprocessing and imposes a burdensome workload on feature computing, resulting in high energy consumpti

Jiatong Xia, Lingqiao Liu

We introduce a novel, training-free system for reconstructing, understanding, and rendering 3D indoor scenes from a sparse set of unposed RGB images. Unlike traditional radiance field approaches that require dense views and per-scene optimization, our pipeline achieves high-fidelity results without any training or pose preprocessing. The system integrates th

Jiang-He Qiao, Jian-Yu Liu, Yan-Ting Feng, Feng-Lan Shao

Based on nucleons ($p$, $n$) and hyperons ($\Lambda$, $\Omega^-$) formed at kinetic freeze-out from a quark combination model, we systematically study the production of light nuclei and hyper-nuclei in the hadronic coalescence picture. We present the analytical formula of the nucleus momentum distribution of two-body coalescence and that of three-body coales

Jhe-Jia Wu, Tian-Li Yan, Ting-Li Chen

Estimating ballpark effects and team defense in baseball is challenging because batted-ball outcomes are influenced by multiple factors, including contact quality, ballpark environment, defensive performance, and random variation. In this study, we propose a simple and interpretable framework based on Total Bases Residuals (TBR). Using Statcast data from 201

Yukio-Pegio Gunji, Yoshihiko Ohzawa, Yuki Tokuyama, Yu Huang

Contextuality is widely regarded as a hallmark of quantum information, yet its structural origin is often obscured by probabilistic or operational formulations. In this work, we show that non-distributive orthomodular structure need not be postulated, but arises canonically as a left adjoint from classical Boolean contexts. We introduce a gluing functor that

Leonid Ugadiarov, Yuri Kuratov, Aleksandr Panov, Alexey Skrynnik

Neural tree search is a powerful decision-making algorithm widely used in complex domains such as game playing and model-based reinforcement learning. Recent work has applied AlphaZero-style tree search to enhance the reasoning capabilities of Large Language Models (LLMs) during inference, but we find that this approach suffers from a scaling failure: on GSM

Rahul D Ray

Epistemic intelligence requires machine learning systems to recognise the limits of their own knowledge and act safely under uncertainty, especially when faced with unknown unknowns. Existing uncertainty quantification methods rely on a single signal such as confidence or density and fail to detect diverse structural anomalies. We introduce SPECTRE-G2, a mul

Duc Duy Tran, Cedric Mannequin, Aboulaye Traore, Masahiro Sasaki

We report the first plasma atomic layer etching (ALE) process for diamond using a cyclic plasma sequence composed of two separated steps: oxygen surface modification and krypton ion removal. The process is implemented in an inductively coupled plasma reactor using alternating O$_2$ plasma exposure and low-energy Kr ion bombardment. This cyclic process exhibi

Filip Noveski, Elena Hadzieva

For the purposes of electric circuit simulation, we consider an iterative simulation model based on solving systems of linear equations by Gauss-Jordan elimination (GJE) for individual moments in time. To accelerate the simulation, we propose two independent novel approaches: a parallel GJE algorithm and partial system reduction prior to the start of iterati

Suining Xiong, Wenwen Zou, Pingwen Zhang, Kai Jiang

We present a unified theoretical and computational framework that bridges mathematical quasiperiodicity with classical crystallographic models. Based on a rigorous cut-and-projection construction, the proposed proximal coincidence point set (PCPS) theory extends the classical coincidence site lattice model and further incorporates physically motivated pertur

Karin Baur, Anna Felikson, Deepanshu Prasad, Pavel Tumarkin

We analyse the growth coefficients of infinite frieze patterns arising from cluster algebras using cluster modular groups and cluster categories. For a fixed cluster category of affine type, we prove that the collection of infinite frieze patterns given by both the homogeneous and non-homogeneous stable tubes all have the same growth coefficients. We also de

Eva A. Gallardo-Gutiérrez, F. Javier González-Doña

In this work, we prove that linear bounded operators $T$ on a Banach space $X$ allowing spectral cuts along rectifiable Jordan curves meeting their spectrum are related to classes of operators admitting an unconventional functional calculus. We identify several such classes and address the consequences regarding the existence of non-trivial closed invariant

Zihui Chen, Yuling Wang, Pengfei Jiao, Kai Wu

Text-attributed graphs (TAGs) enhance graph learning by integrating rich textual semantics and topological context for each node. While boosting expressiveness, they also expose new vulnerabilities in graph learning through text-based adversarial surfaces. Recent advances leverage diverse backbones, such as graph neural networks (GNNs) and pre-trained langua

Ang A. Li

Changepoint detection is increasingly applied to ecological time series, yet statistical power at the short series lengths typical of monitoring (10-50 observations) is rarely assessed. We present a simulation-based power analysis for BIC-based Binary Segmentation across 108 combinations of series length, effect size, and number of changepoints. BIC achieves

Tianhao Ma, Ximing Li, Changchun Li, Renchu Guan

Learning from Label Proportions (LLP) is a weakly supervised problem in which the training data comprise bags, that is, groups of instances, each annotated only with bag-level class label proportions, and the objective is to learn a classifier that predicts instance-level labels. This setting is widely applicable when privacy constraints limit access to inst

Feng Liu, Jian Xu, Xin Cui, Xinghao Wang

Inferring physical mechanisms that govern earthquake sequences from geophysical observations remains a challenging task, particularly across tectonically distinct environments where similar seismic patterns can reflect different underlying processes. Current seismological processing and interpretation rely heavily on experts' choice of parameters and the syn

Tomás J. Ferreyra Hauchar, Detlef Lohse, Sander G. Huisman

The salinity of water and the slope of ice significantly influence the melt rate and surface morphology of ice, both highly relevant in the context of glacier and iceberg melting in oceanic environments. In this study, we conducted experiments on vertical and sloped ice blocks melting in quiescent saline water. Through the use of fringe projection profilomet

Octavian Untila

An autonomous AI ecosystem (SUBSTRATE S3), generating product specifications without explicit instructions about formal methods, independently proposed the use of Z3 SMT solver across six distinct domains of AI safety: verification of LLM-generated code, tool API safety for AI agents, post-distillation reasoning correctness, CLI command validation, hardware

Fangyuan Li, Pengfei Li, Shijie Wang, Junqi Gao

Recent progress in reinforcement learning with verifiable rewards (RLVR) offers a practical path to self-improvement of language models, but existing methods face a key trade-off: endogenous self-play can drift over iterations, while corpus-grounded approaches rely on curated data environments. We present \textbf{WIST}, a \textbf{W}eb-grounded \textbf{I}tera

Robert Krauthgamer, Nir Petruschka

The Nearest Neighbor Search (NNS) problem asks to design a data structure that preprocesses an $n$-point dataset $X$ lying in a metric space $\mathcal{M}$, so that given a query point $q \in \mathcal{M}$, one can quickly return a point of $X$ minimizing the distance to $q$. The efficiency of such a data structure is evaluated primarily by the amount of space

Yu-Feng Mao, Shicheng Ma, Yong Xu

Despite long-standing theoretical interest, the chiral spin liquid, a topologically ordered phase, has yet to be observed experimentally. Here we surprisingly find its emergence in an experimentally realized dipolar $\text{XY}$ model when Rydberg atoms are arranged in a breathing kagome lattice. Using the infinite density matrix renormalization group, we num

Huyuan Chen, Rui Chen, Daniel Hauer

In this paper, we introduce the conformal fractional--logarithmic Laplacian on the unit sphere, defined as the derivative of the conformal fractional Laplacian with respect to the order parameter \(s\in(0,1)\). We investigate its fundamental analytic and spectral properties, including its relation to the conformal logarithmic Laplacian, its spectral represen

Peihan Ye, Alfreds Lapkovskis, Alaa Saleh, Qiyang Zhang

The computational demands of modern AI services are increasingly shifting execution beyond centralized clouds toward a computing continuum spanning edge and end devices. However, the scale, heterogeneity, and cross-layer dependencies of these environments make resilience difficult to maintain. Existing fault-management methods are often too static, fragmente

MD Ruiz-Medina, AE Madrid, A Torres-Signes, JM Angulo

This paper proposes a new formulation of functional Gaussian Process regression in manifolds, based on an Empirical Bayes approach, in the spatiotemporal random field context. We apply the machinery of tight Gaussian measures in separable Hilbert spaces, exploiting the invariance property of covariance kernels under the group of isometries of the manifold. T

Masaki Tsutsumi, Takuya Kiyokawa, Gen Sako, Kensuke Harada

Robotic disassembly of complex mating components often renders pinch grasping infeasible, necessitating multi-fingered enveloping grasps. However, visual occlusions and geometric constraints complicate teaching appropriate grasp motions when relying solely on 2D camera feeds. To address this, we propose an affordance-guided teleoperation method that pre-gene

Jeffrey Chen, Rohan Chandra

Robots operating in dynamic, unstructured environments must balance safety and efficiency under potentially limited sensing. While control barrier functions (CBFs) provide principled collision avoidance via safety filtering, their behavior is often governed by fixed parameters that can be overly conservative in benign scenes or overly permissive near hazards

Chio Chon Kit

This paper constructs a rigorous mathematical framework for investigating laminar-turbulent transition induced by weak singularities of incompressible Navier-Stokes (NS) equations. By integrating the energy identity of Leray weak solutions with the singularity criterion $\left\lVert \boldsymbol{u} \right\rVert_{H_0^1(\Omega)}\to0$, a closed analytical form o

Zhuojie Yang, Wentao Wan, Keze Wang

Training large language models (LLMs) with synthetic reasoning data has become a popular approach to enhancing their reasoning capabilities, while a key factor influencing the effectiveness of this paradigm is the quality of the generated multi-step reasoning data. To generate high-quality reasoning data, many recent methods generate synthetic reasoning path

Wenjin Hou, Xiaoxiao Sun, Hehe Fan

Recent advances in zero-shot learning (ZSL) have demonstrated the potential of generative models. Typically, generative ZSL synthesizes visual features conditioned on semantic prototypes to model the data distribution of unseen classes, followed by training a classifier on the synthesized data. However, the synthesized features often remain task-agnostic, le

Zhirui Shen, Bin Wang

This paper investigates the equations of motion for a relativistic charged particle in a general magnetic field. By reformulating the dynamics in four-dimensional spacetime and separating the linear and nonlinear parts, we construct an explicit symmetric exponential integrator based on Lie splitting. Rigorous analysis establishes its unconditional stability

Pengxiang Cai, Mengyang Li

Diffusion-based text-to-image generation has advanced significantly, yet customizing scenes with multiple distinct subjects while maintaining fine-grained control over their interactions remains challenging. Existing methods often struggle to provide explicit user-defined control over the compositional structure and precise spatial relationships between subj

Yu-Wen Tseng, Xingyi Zheng, Ya-Chen Wu, I-Bin Liao

Test-time adaptation (TTA) adapts pre-trained models to distribution shifts at inference using only unlabeled test data. Under the Practical TTA (PTTA) setting, where test streams are temporally correlated and non-i.i.d., memory has become an indispensable component for stable adaptation, yet existing methods universally store amples in a single unstructured

Zhiyan Cao, Zhengxi Wu, Yiwei Wang, Pei-Hsuan Lin

Cardiac ultrasound diagnosis is critical for cardiovascular disease assessment, but acquiring standard views remains highly operator-dependent. Existing medical segmentation models often yield anatomically inconsistent results in images with poor textural differentiation between distinct feature classes, while autonomous probe adjustment methods either rely

E. Alkin, A. Miroshnikov, A. Skopenkov

In this expository note we present an elementary direct rigorous definition and the simplest properties of the winding number. This definition is simpler than the one given in some textbooks. We show how to compute the winding number easily: using additivity or counting the (signed) intersection points. In the language of the winding number, we present an el

Ao Ding, Aoxing Zhang, Tzung-May Fu, Yuanlong Huang

Current operational air quality forecasts are computationally expensive, sensitive to errors in physics and emissions, and often neglect weather-related uncertainty. To address these limitations, we present AirFusion, a hybrid, diffusion-based framework that synergistically integrates knowledge from chemical transport models with real-world observational con

Tina Šfiligoj, Renzo Massobrio, Oded Cats

The introduction of network science approaches into public transport research has seen great advances in the past 15 years. However, it has become apparent that monolayer networks are often not sufficient to model and analyse real-world systems in sufficient detail. In the last decade, the theory of multilayer networks has proven to be an invaluable tool in

Bo Li, Tingting Bao, Lingling Zhang, Weiping Fu

Diffusion models have achieved impressive performance on multi-focus image fusion (MFIF). However, a key challenge in applying diffusion models to the ill-posed MFIF problem is that defocus blur can make common symmetric geometric structures (e.g., textures and edges) appear warped and deformed, often leading to unexpected artifacts in the fused images. Ther

Linquan Shao, Haoyu Yan, Yingjun Chen, Jiaxin Pi

In this paper, the global $\chi^2$ function for the simultaneous fitting of correlated energy-dependent cross sections is constructed, where the correlations between the measured cross sections of different processes and/or at different center-of-mass energy points, as well as the contributions from the integrated luminosity measurement and the center-of-mas

Shengjia Wang, Tian-Nuo Li, Tonghua Liu, Guo-Hong Du

We present a model-independent reconstruction of the quintessence scalar field's dynamics-both its potential and kinetic energy-directly from the latest cosmological observations. Our analysis combines DESI DR2 baryon acoustic oscillation measurements with the Pantheon plus Type Ia supernova compilation, employing Gaussian process with four distinct covarian

Masaru Ikehata

The Probe Method is an analytical reconstruction scheme for inverse obstacle problems utilizing the Dirichlet-to-Neumann map associated with the governing partial differential equation. It consists of two distinct parts: Side A and Side B. Both are based on the indicator sequence which is calculated from the Dirichlet-to-Neumann map acting on "needle-like" s

Fanxing Li, Fangyu Sun, Tianbao Zhang, Shuyu Wu

First-order reinforcement learning with differentiable simulation is promising for quadrotor control, but practical progress remains fragmented across task-specific settings. To support more systematic development and evaluation, we present a unified differentiable framework for multi-task quadrotor control. The framework is wrapped, extensible, and equipped

Xiaoning Feng, Arman Nejad, David P. Tew

We develop a time-dependent, grid-based framework for simulating infrared spectra that is specifically designed for quantum computers. The proposed circuit employs a probabilistic strategy for applying the non-unitary dipole operator and an Split Operator-Quantum Fourier Transform time evolution scheme. Using a vibrational model of the water molecule as a te

Panu Lahti, Yuxin Li, Khanh Nguyen

We establish a weak Harnack inequality for nonlocal $W^{s,1}$-subminimizers in a complete, connected, doubling metric measure space where $0<s<1$. As a corollary, we prove that $W^{s,1}$-subminimizers are semicontinuous, up to a suitable choice of pointwise representative. We then prove \emph{Cartan-type properties} for $W^{s,1}$-superminimizers. The theory

Yajun Zhang, Devesh R. Kripalani, Xu He, Konstantin Shapovalov

Spin density waves (SDWs) represent a fundamental paradigm of spatially modulated order in condensed matter systems, yet their electrical and mechanical analogues polarization and strain density waves (PDWs and StDWs) have remained elusive as equilibrium phases. Here, we introduce a general, symmetry-driven strategy to unlock static PDWs and StDWs in perovsk

Florin Adrian Chitan

Deterministic pre-execution safety gates evaluate whether individual agent actions are compatible with their assigned roles. While effective at per-action authorization, these systems are structurally blind to distributed attacks that decompose harmful intent across multiple individually-compliant steps. This paper introduces Session Risk Memory (SRM), a lig

Mikhail Kornev

We introduce a new class of formal group laws whose modulus square construction yields Buchstaber's family of polynomials. This class is related to, but does not coincide with, the family of formal group laws associated with the Krichever genus. We compute the values of the corresponding Hirzebruch genus on theta divisors and complex projective spaces, descr

Yuhao Pan, Wenchao Xu, Fushuo Huo, Haozhao Wang

Tor is a low-latency anonymous communication network that protects user privacy by encrypting website traffic. However, recent website fingerprinting (WF) attacks have shown that encrypted traffic can still leak users&#39; visited websites by exploiting statistical features such as packet size, direction, and inter-arrival time. Most existing WF attacks form

Mounir Nisse

The topology of amoebas of complex algebraic hypersurfaces is deeply connected to the combinatorics of the Newton polytope and the convex geometry of the Ronkin function. A long-standing conjecture of Passare and Rullgard asserts that the amoeba of a maximally sparse Laurent polynomial, whose support consists exactly of the vertices of its Newton polytope, m

Xiaoshan Wu, Xiaoyang Lyu, Yifei Yu, Bo Wang

Dense semantic segmentation in dynamic environments is fundamentally limited by the low-frame-rate (LFR) nature of standard cameras, which creates critical perceptual gaps between frames. To solve this, we introduce Anytime Interframe Semantic Segmentation: a new task for predicting segmentation at any arbitrary time using only a single past RGB frame and a

Shanmukha Vellamcheti, Uday Kiran Kothapalli, Disharee Bhowmick, Sathyanarayanan N. Aakur

Multimodal large language models (MLLMs) perform strongly on isolated spatial tasks, but whether their predictions remain persistent and mutually coherent across viewpoints and competing scenes is unclear. We formalize this behavioral property as spatial-state integrity and introduce CVT-Bench, a factorial diagnostic suite spanning two domains (CVT-Synthetic

Evgeny Korotyaev

We consider the scattering for the operator $H=H_o+V$, where the unperturbed operator $H_o$ is not assumed to be elliptic and the potential $V$ is anisotropic. Under some conditions on $H_o$ and $V$ we show that the wave operators for $H_o, H$ exist and are complete, $H$ has no singular continuous spectrum and the eigenvalues of $H$ can accumulate only to ze

Shih-Wen Liu, Yen-Chang Chen, Wei-Ta Chu, Fu-En Yang

Multi-task learning (MTL) aims to enable a single model to solve multiple tasks efficiently; however, current parameter-efficient fine-tuning (PEFT) methods remain largely limited to single-task adaptation. We introduce \textbf{Free Sinewich}, a parameter-efficient multi-task learning framework that enables near-zero-cost weight modulation via frequency swit

Xiong Jiang, Hai Lei, Hao-Yi Huang, Wei Zhang

We report the first detection of GeV gamma-ray flaring activity from the compact symmetric object (CSO)-like source 4C 76.03, based on 17 years of Fermi-LAT observations. Its long-term, time-averaged gamma-ray properties are consistent with the 4FGL-DR4 catalog. However, a time-resolved analysis with 100-day binning reveals two prominent flares occurring on

Yicheng Di

In the domain of consumer electronics, personalized sequential recommendation has emerged as a central task. Current methodologies in this field are largely centered on modeling user behavior and have achieved notable performance. Nevertheless, the inherent quadratic computational complexity typical of most existing approaches often leads to inefficiencies t

Yanxian Tao, Lingyun Wan, Jie Liu

The variational quantum eigensolver (VQE) has emerged as one of the leading quantum algorithms for solving electronic structure problems on near-term noisy intermediate-scale quantum devices. However, its practical application to quantum chemistry remains challenging due to the limited coherence time, imperfect quantum gate fidelity, and the large number of

Long Xu, Junping Guo, Jianbo Zhao, Jianbo Lu

Molecular property prediction constitutes a cornerstone of drug discovery and materials science, necessitating models capable of disentangling complex structure-property relationships across diverse molecular modalities. Existing approaches frequently exhibit entangled representations--conflating structural, chemical, and functional factors--thereby limiting

Kazi Ababil Azam, Imtiaz Karim, Dipto Das

Romantic AI chatbots have quickly attracted users, but their emotional use raises concerns about privacy and safety. As people turn to these systems for intimacy, comfort, and emotionally significant interaction, they often disclose highly sensitive information. Yet the privacy implications of such disclosure remain poorly understood in platforms shaped by p

Xu Li, Yi Zheng, Yuxuan Liang, Zhe Liu

Large Vision-Language Models (LVLMs) rely on dense visual tokens to capture fine-grained visual information, but processing all these tokens incurs substantial computational and memory overhead during inference. To address this issue, we propose ResPrune, a training-free visual token pruning framework that enables efficient LVLM inference by selecting a comp

Bowen Jing, Ruiyang Hao, Weitao Zhou, Haibao Yu

Generating safety-critical driving scenarios requires understanding why dangerous interactions arise, rather than merely forcing collisions. However, existing methods rely on heuristic adversarial agent selection and unstructured perturbations, lacking explicit modeling of interaction dependencies and thus exhibiting a realism--adversarial trade-off. We pres

Kazım Yavuz Ekşi, İrem Bakır

Central compact objects are radio-quite young neutron stars associated with supernova remnants. They have relatively small dipole fields, $B_{\rm p} \sim 10^{10}\,{\rm G}$ as inferred from their spin parameters. X-ray observations and theoretical arguments imply the presence of stronger internal magnetic fields. We argue that the dipole fields of these objec

Hao Luo, Zhiyuan Zhai, Qianli Zhou, Jun Qi

Quantum federated learning (QFL) has recently emerged as a promising paradigm for privacy-preserving collaborative learning, yet most existing studies focus on horizontal federated learning and ignore the vertical federated learning (VFL), where parties hold complementary features of aligned samples. In this work, we propose Evidential Quantum Vertical Feder

Junyan Chu

Saito's criterion is a foundational result that algebraically characterizes free hyperplane arrangements via the determinant of a square matrix of logarithmic derivations. It is natural to ask whether this criterion can be generalized to the non-free setting. To address this, we formulate a general problem concerning the maximal minors of a $p \times \ell$ (

He Wang, Tianyang Xu, Zhangyong Tang, Xiao-Jun Wu

Due to the limited availability of paired multi-modal data, multi-modal trackers are typically built by adopting pre-trained RGB models with parameter-efficient fine-tuning modules. However, these fine-tuning methods overlook advanced adaptations for applying RGB pre-trained models and fail to modulate a single specific modality, cross-modal interactions, an

Yasushi Homma, Natsuki Imada, Soma Ohno

We define higher spin Killing spinors on Riemannian spin manifolds in arbitrary dimension and study them in detail in dimension three. We prove a rigidity result for 3-dimensional manifolds admitting higher spin Killing spinors and give expressions for higher spin Killing spinors on the 3-sphere and the 3-hyperbolic space explicitly. We also investigate the

Koushiki, W. Piechocki, G. Plewa

We examine the existence of an apparent horizon in the collapse of an isolated dust cloud using expansion functions. Our results indicate that in the region of spacetime far away from the gravitational singularity, the considered system has a horizon, in which case the singularity is covered. Using this method may have limited applicability in the neighbourh

Haidong Wang, Songhan Zhao, Bo Gu, Shimin Gong

In this paper, we explore a joint source and reconfigurable intelligent surface (RIS)-assisted channel encoding (JSRE) framework for multi-user semantic communications, where a deep neural network (DNN) extracts semantic features for all users and the RIS provides channel orthogonality, enabling a unified semantic encoding-decoding design. We aim to maximize

Tasmay Pankaj Tibrewal, Pritish Saha, Ankit Meda, Kunal Singh

Transformers lack an explicit architectural mechanism for storing and organizing knowledge acquired during training. We introduce learnable sparse memory banks: a set of latent tokens, randomly initialized and trained end-to-end, that transformer layers query via cross-attention to retrieve stored knowledge. To scale memory capacity without prohibitive atten

Dina Salama, Mohamed Mahmoud, Nourhan Bayasi, David Liu

Thyroid ultrasound is the first-line exam for assessing thyroid nodules and determining whether biopsy is warranted. In routine reporting, radiologists produce two coupled outputs: a nodule contour for measurement and a TI-RADS risk category based on sonographic criteria. Yet both contouring style and risk grading vary across readers, creating inconsistent s

Smitha Muthya Sudheendra, Jaideep Srivastava

Human annotation is central to NLP evaluation, yet subjective tasks often exhibit substantial variability across annotators. While large language models (LLMs) can provide structured reasoning to support annotation, their influence on human annotation behavior remains underexplored. We introduce \textbf{ReasonScaffold}, a scaffolded reasoning annotation prot

Songhan Zhao, Shimin Gong, Bo Gu, Zehui Xiong

This paper explores a reconfigurable intelligent surface (RIS)-assisted and semantic-aware wireless network, where multiple semantic users (SUs) transmit semantic information to an access point (AP) using the non-orthogonal multiple access (NOMA) method. The RIS reconfigures channel conditions, while semantic extraction reshapes traffic demands, providing en

Songhan Zhao, Shimin Gong, Bo Gu, Hongyang Du

In this paper, we investigate a generative artificial intelligence (GAI)-assisted semantic communication framework for non-orthogonal multiple access (NOMA)-based image transmissions. Semantic users (SUs) extract cross-modal semantic features from the raw images, which are then used for image recovery by leveraging a GAI model. The GAI enhances the generaliz

Vivek Shripad Borkar

Based on some recent work of the author on stochastic approximation in non-markovian environments, the situation when the driving random process is non-ergodic in addition to being non-markovian is considered. Using this, we propose an analytic framework for understanding transformer based learning, specifically, the `attention' mechanism, and continual lear

Lingling Zhang, Pengpeng Qiao, Zhiwei Zhang, Ye Yuan

Temporal Graph Neural Networks (TGNs) achieve state-of-the-art performance on dynamic graph tasks, yet existing systems focus exclusively on accelerating training -- at inference time, every new edge triggers $O(|V|)$ embedding updates even though only a small fraction of nodes are affected. We present \textbf{StreamTGN}, the first streaming TGN inference sy

Flemming Holtorf, Sungho Shin

We present an approximate dynamic programming framework for designing degradation-aware market participation policies for battery energy storage systems. The approach employs a tailored value function approximation that reduces the state space to state of charge and battery health, while performing dynamic programming along a pseudo-time axis encoded by stat

Yifan Wang, Jeonghun Lee, Suncica Canic

We present a priori error analysis for a fully discrete, parallelizable, explicit loosely coupled scheme for the time-dependent Stokes-Biot problem. The method decouples the fluid and poroelastic subproblems in a fully explicit fashion, allowing each problem to be solved independently at each time step, with a consistent treatment of the interface conditions

Zeyang Sun, Shuai Han, Chenyu Wu, Sai Xu

Information-bearing reconfigurable intelligent surfaces (IB-RIS) provide a promising solution to self-sustainable and green communications by harvesting ambient radio frequency energy while embedding information via passive reflection. This paper investigates a self-sustainable IB-RIS (SIB-RIS)-assisted non-orthogonal multiple access (NOMA) network operating

Bahram Mohammadi, Yanqiu Wu, Vu Minh Hieu Phan, Sam White

Accurate brain tumor segmentation from MRI scans is critical for diagnosis and treatment planning. Despite the strong performance of recent deep learning approaches, two fundamental limitations remain: (1) the lack of reliable uncertainty quantification in single-model predictions, which is essential for clinical deployment because the level of uncertainty m

Qifan Li, Xingyu Zhou, Jinhua Zhang, Weiyi You

Latent diffusion models have emerged as the dominant framework for high-fidelity and efficient image generation, owing to their ability to learn diffusion processes in compact latent spaces. However, while previous research has focused primarily on reconstruction accuracy and semantic alignment of the latent space, we observe that another critical factor, ro