arXiv preprints from January 1, 2026 through September 25, 2026 — 14:49:11 EST
Large foundation models are integrated into Computer Use Agents (CUAs), enabling autonomous interaction with operating systems through graphical user interfaces (GUIs) to perform complex tasks. This autonomy introduces serious security risks: malicious instructions or visual prompt injections can trigger unsafe reasoning and cause...
In this paper, we focus on the sub-wavelength resonances in two-dimensional elastic media characterized by high contrasts in both Lamé parameters and density. Our contributions are fourfold. First, it is proved that the operator $\hat{\mathbf{S}}_{\partial D}^ω$, which serves as a leading order approximation to $\mathbf{S}_{\partial...
Total-body PET/CT enables system-wide molecular imaging, but heterogeneous anatomical and metabolic signals, approximately 2 m axial coverage, and structured radiology semantics challenge existing medical AI models that assume single-modality inputs, localized fields of view, and coarse image-text alignment. We introduce SDF-HOLO...
The COSINE-100U (upgrade) will enhance the sensitivity of the COSINE-100 dark matter search by operating the detector array immersed in liquid scintillator (LS) at $-30^oC$. To validate the detector design for these conditions, we constructed a module using the COSINE-100U encapsulation and performed a dedicated long-term stability...
The Lloyd Theorem of (Solé, 1989) is combined with the Schwartz-Zippel Lemma of theoretical computer science to derive non-existence results for perfect codes in the Lee metric, NRT metric, mixed Hamming metric, and for the sum-rank distance. The proofs are based on asymptotic enumeration of integer partitions. The framework is the...
Liability sharing and staffing jointly determine service quality in AI-assisted online medical consultation, yet their interaction is rarely examined in an integrated framework linking contracts to congestion via physician responses. This paper develops a Stackelberg queueing model where the platform selects a liability share and a...
The magnetic-field-induced correlated Chern insulator (CCI) states in magic-angle twisted bilayer graphene (MATBG) have been intensively studied in experiments, but a simple and clear understanding of their origin is still lacking. Here, we propose a unified theoretical framework for the CCI states in MATBG that successfully explains...
Nonlinear aggregation is central to modern distributed systems, yet its privacy behavior is far less understood than that of linear aggregation. Unlike linear aggregation where mature mechanisms can often suppress information leakage, nonlinear operators impose inherent structural limits on what privacy guarantees are theoretically...
We study the escape of particles in the lemon billiard, a two-parameter family of billiard systems defined by the intersection of two identical circles. Using numerical simulations, we explore how the survival probability depends on the position and size of the hole, as well as on the billiard shape parameter. We find that the...
We report the discovery of an exceptionally eccentric binary system, BE Lyncis (BE Lyn), which might host a compact companion with mass $\gtrsim 2.5~M_{\odot}$. By combining TESS photometry with an extensive set of times of maximum light spanning 39 years, we identify BE Lyn as a high-amplitude $δ$ Scuti star in a binary with an...
The weighted-Hamming metric generalizes the Hamming metric by assigning different weights to blocks of coordinates. It is well-suited for applications such as coding over independent parallel channels, each of which has a different level of importance or noise. From a coding-theoretic perspective, the actual error-correction...
Let $X_1,\ldots,X_M$ and $Y_1,\ldots,Y_N$ be independent zero mean normal random variables with variances $σ_{X_i}^2$, $i=1,\ldots,M$, and $σ_{Y_j}^2$, $j=1,\ldots,N$, respectively, and let $X=X_1\cdots X_M$ and $Y=Y_1\cdots Y_N$. In this paper, we derive the exact probability density function of the ratio $X/Y$. We apply this formula...
Edge learning facilitates ubiquitous intelligence by enabling model training and adaptation directly on data-generating devices, thereby mitigating privacy risks and communication latency. However, the high computational and energy overhead of on-device training hinders its deployment on battery-powered mobile systems with strict...
Multi-Agent Systems (MAS) offer a powerful paradigm for solving complex problems, yet their performance is critically dependent on the design of their underlying collaboration topology. As MAS become increasingly deployed in web services (e.g., search engines), designing adaptive topologies for diverse cross-domain user queries...
Long Chain-of-Thought (LCoT), achieved by Reinforcement Learning with Verifiable Rewards (RLVR), has proven effective in enhancing the reasoning capabilities of Large Language Models (LLMs). However, reasoning in current LLMs is primarily generated as plain text, where performing semantic evaluation on such unstructured data creates a...
In autonomous driving, multi-modal perception tasks like 3D object detection typically rely on well-synchronized sensors, both at training and inference. However, despite the use of hardware- or software-based synchronization algorithms, perfect synchrony is rarely guaranteed: Sensors may operate at different frequencies, and...
We introduce Being-H0.5, a foundational Vision-Language-Action (VLA) model designed for robust cross-embodiment generalization across diverse robotic platforms. While existing VLAs often struggle with morphological heterogeneity and data scarcity, we propose a human-centric learning paradigm that treats human interaction traces as a...
In this article we provide Bernstein type gradient estimates for two system of local weighted heat type equations with potentials on a weighted Riemannian manifold. We derive all possible cases considering linear potential, exponential potential, combining with static manifold and evolving manifold. This work partially resolved the...
Recent evidence from the ATLAS Collaboration for the rare decay $H \rightarrow μ^{+}μ^{-}$ provides a unique window into the Higgs boson's coupling to second-generation fermions. In this work, we investigate how this signal can probe physics beyond the Standard Model by computing the branching ratios $B(H_{i} \rightarrow μ^{+}μ^{-})$...
The advent of Retrieval-Augmented Generation (RAG) has significantly enhanced the ability of Large Language Models (LLMs) to produce factually accurate and up-to-date responses. However, the performance of a RAG system is not determined by a single component but emerges from a complex interplay of modular choices, such as embedding...
Generative models for financial time series often create data that look realistic and even reproduce stylized facts such as fat tails or volatility clustering. However, these apparent successes break down under trading backtests: models like GANs or WGAN-GP frequently collapse, yielding extreme and unrealistic results that make the...
In blockchain systems operating under the Proof-of-Stake (PoS) consensus mechanism, fairness in transaction processing is essential to preserving decentralization and maintaining user trust. However, with the emergence of Maximal Extractable Value (MEV), concerns about economic centralization and content manipulation have intensified....
The accelerating growth of the scientific literature makes it increasingly difficult for researchers to track new advances through manual reading alone. Recent progress in large language models (LLMs) has therefore spurred interest in autonomous agents that can read scientific papers and extract task-relevant information. However,...
This paper develops a guidance control law based on a parametric Guiding Vector Field (GVF) and integrates it with a state-of-the-art acceleration and attitude control architecture for tailsitters. The resulting framework enables a direct comparison between traditional trajectory-tracking guidance and GVF-based path-following guidance...
Protecting the intellectual property of large language models requires robust ownership verification. Conventional backdoor fingerprinting, however, is flawed by a stealth-robustness paradox: to be robust, these methods force models to memorize fixed responses to high-perplexity triggers, but this targeted overfitting creates...