arXiv preprints from January 1, 2026 through September 24, 2026 — 07:53:53 EST
This article presents a data-driven system-level analysis of multi-layer 6G networks operating in the upper mid-band (FR3: 7-24 GHz). Unlike most prior studies based on 3rd Generation Partnership Project (3GPP) templates, we leverage real-world deployment and traffic data from a commercial 4G/5G network in China to evaluate practical...
Chain-of-Thought (CoT) reasoning empowers Large Language Models (LLMs) with remarkable capabilities but typically requires prohibitive parameter scales. CoT distillation has emerged as a promising paradigm to transfer reasoning prowess into compact Student Models (SLMs), but existing approaches often rely on a solitary teacher,...
A fundamental computational task in probabilistic programming is to infer a program's output (posterior) distribution from a given initial (prior) distribution. This problem is challenging, especially for expressive languages that feature loops or unbounded recursion. While most of the existing literature focuses on statistical...
We consider reachable sets of switching systems, which are families of linear ODE $\dot x(t) \, = \, \, A(t) \, x(t)$ with a function $A(\cdot)$ taking values on a given compact set of $d\times d$ matrices. An eigenset is a compact set $M \ne \{0\}$ that possesses the following property: the closure of the set of points reachable by...
Neural networks are predominantly trained using gradient-based methods, yet in many applications their final predictions remain far from the accuracy attainable within the model's expressive capacity. We introduce Linearized Subspace Refinement (LSR), a general and architecture-agnostic framework that exploits the Jacobian-induced...
In this expository paper, we review the history and the recent breakthroughs in the spectral theory of large volume hyperbolic surfaces. More precisely, we focus mostly on the investigation of the first non-trivial eigenvalue $λ_1$ and its possible behaviour in the large volume regime.
Hyperspectral image (HSI) restoration is a fundamental challenge in computational imaging and computer vision. It involves ill-posed inverse problems, such as inpainting and super-resolution. Although deep learning methods have transformed the field through data-driven learning, their effectiveness hinges on access to meticulously...
Image Dehazing (ID) aims to produce a clear image from an observation contaminated by haze. Current ID methods typically rely on carefully crafted priors or extensive haze-free ground truth, both of which are expensive or impractical to acquire, particularly in the context of scientific imaging. We propose a new unsupervised learning...
Component systems - ensembles of realizations built from a shared repertoire of modular parts - are ubiquitous in biological, ecological, technological, and socio-cultural domains. From genomes to texts, cities, and software, these systems exhibit statistical regularities that often meet the "bona fide" requirements of laws in the...
We present an experimental study on steady gravity currents advancing along a heated wall. The current is generated by a mixture of air and carbon dioxide continuously supplied at the channel inlet. To have a complete point-wise characterization of the flow, simultaneous high-frequency measurements of two velocity components, CO_2...
Two applications of gates from the B gate equivalence class can generate all two-qubit gates. This local equivalence class is invariant under the mirror (multiplication with the SWAP gate) operation, inverse (Hermitian conjugate) operation, and the combined inverse and mirror operations. The last two symmetries are associated with the...
Ongoing advances in scanning probe microscopy techniques are continually expanding the possibilities for nanoscale characterization and correlated studies of functional materials. Here, we demonstrate how a recent extension of piezoresponse force microscopy (PFM), known as interferometric vector PFM, can be utilized for simultaneously...
Large language models (LLMs) have shown strong capabilities in multi-step decision-making, planning and actions, and are increasingly integrated into various real-world applications. It is concerning whether their strong problem-solving abilities may be misused for crimes. To address this gap, we propose VirtualCrime, a sandbox...
Quasi-poloidal (QP) magnetic fields have desirable properties for confining plasma: no radial drift of guiding centres (with positive implications for neoclassical transport), zero Pfirsch-Schlüter current, and a lower level of damping for poloidal flows. Despite their attractive properties, QP fields are not amenable to the near-axis...
This paper presents a novel cross-modal visuo-tactile perception framework for the 3D shape reconstruction of deformable linear objects (DLOs), with a specific focus on cables subject to severe visual occlusions. Unlike existing methods relying predominantly on vision, whose performance degrades under varying illumination, background...
The luminous accreting pulsar SMC X-1 is an appropriate target to explore the accretion dynamics. SMC X-1 shows unique quasi-periodic flux variability of 40-65$\,$days known as superorbital modulation. To constrain the accretion structure of SMC X-1 based on timing and spectral study, we have analysed X-ray data of SMC X-1 observed by...
Residues appear naturally in various questions in complex and algebraic geometry: interpolation, duality, representation problems, and obstructions. The first global vanishing result in the projective plane, known as the Euler-Jacobi theorem, was established by Jacobi in 1835. In the toric case, the input is a system of n Laurent...
Achieving human-level performance in Vision-and-Language Navigation (VLN) requires an embodied agent to jointly understand multimodal instructions and visual-spatial context while reasoning over long action sequences. Recent works, such as NavCoT and NavGPT-2, demonstrate the potential of Chain-of-Thought (CoT) reasoning for improving...
Marine biodiversity monitoring requires scalability and reliability across complex underwater environments to support conservation and invasive-species management. Yet existing detection solutions often exhibit a pronounced deployment gap, with performance degrading sharply when transferred to new sites. This work establishes the...
Frame sampling is a fundamental component in video understanding and video--language model pipelines, yet evaluating the quality of sampled frames remains challenging. Existing evaluation metrics primarily focus on perceptual quality or reconstruction fidelity, and are not designed to assess whether a set of sampled frames adequately...
Objective: This paper develops a theoretical framework explaining when and why AI explanations enhance versus impair human decision-making. Background: Transparency is advocated as universally beneficial for human-AI interaction, yet identical AI explanations improve decision quality in some contexts but impair it in others. Current...
We propose a Bayesian evidence-based inference framework based on relative belief ratios and apply it to discriminating between one and two incoherent optical point sources using spatial-mode demultiplexing (SPADE). Unlike the Helstrom measurement, SPADE require no collective detection and its optimal for asymptotically large samples....
We present a novel framework for rigid body dynamics in ambient media, such as air or water, enabling accurate motion prediction of objects without requiring computational fluid dynamics simulations. Our method computes the added mass of the fluid and replaces heuristic models for shape-dependent lift and drag with a generalized...
Quantum hypothesis testing concerns the discrimination between quantum states. This paper introduces a novel lower bound for asymmetric quantum hypothesis testing that is based on the Nussbaum-Szkoła mapping. The lower bound provides a unified recovery of converse results across all major asymptotic regimes, including large-,...
Large Language Models (LLMs) are widely integrated into interactive systems such as dialogue agents and task-oriented assistants. This growing ecosystem also raises supply-chain risks, where adversaries can distribute poisoned models that degrade downstream reliability and user trust. Existing backdoor attacks and defenses are largely...