arXiv preprints from January 1, 2026 through October 2, 2026 — 20:07:58 EST
While Large Language Models (LLMs) excel at factual retrieval, they often struggle with the "curse of two-hop reasoning" in compositional tasks. Recent research suggests that parameter-sharing transformers can bridge this gap by forming a "Generalization Circuit" during a prolonged "grokking" phase. A fundamental question arises: Is a...
Extended Reality (XR) affords an enhanced sense of bodily presence that supports experiential modes of comprehension and affective engagement which exceed the possibilities of conventional information delivery. Nevertheless, the psychological processes engendered by XR, and the manner in which these processes inform subsequent...
High-frequency instabilities in crossed-field ($\bm E\times\bm B$) plasmas are widely implicated in anomalous cross-field electron transport in Hall thrusters and related devices. Building on the fully kinetic 3D electrostatic dispersion relations reported by Ducrocq \emph{et al.} and later by Lafleur \emph{et al.}, we provide a...
We investigate the construction of circulant matrices derived from primitive roots over finite fields. Our approach reduces exponential sums to Jacobi sums, thereby establishing explicit connections between character theory and matrix structures. The results provide new insights into the interaction between additive and multiplicative...
V-based Kagome metals exhibit a unique lattice geometry that can give rise to exotic electronic and magnetic phenomena, making them an ideal platform to study the interplay of topology and magnetism. We present a combined thermodynamic and muon spin relaxation ($μ$SR) investigation of single-crystal RV$_{6}$Sn$_{6}$ (R = Tb, Dy, Ho,...
Extended Reality (XR) offers transformative potential for industrial support, training, and maintenance; yet, widespread adoption lags despite demonstrated occupational value and hardware maturity. Organizations successfully implement XR in isolated pilots, yet struggle to scale these into sustained operational deployment, a...
Medical imaging datasets often suffer from class imbalance and limited availability of pathology-rich cases, which constrains the performance of machine learning models for segmentation, classification, and vision-language tasks. To address this challenge, we propose POWDR, a pathology-preserving outpainting framework for 3D MRI based...
Horseshoe mixtures-of-experts (HS-MoE) models provide a Bayesian framework for sparse expert selection in mixture-of-experts architectures. We combine the horseshoe prior's adaptive global-local shrinkage with input-dependent gating, yielding data-adaptive sparsity in expert usage. Our primary methodological contribution is a particle...
This work addresses the fundamental problem of unbounded metric movement costs in bandit online convex optimization, by considering high-dimensional dynamic quadratic hitting costs and $\ell_2$-norm switching costs in a noisy bandit feedback model. For a general class of stochastic environments, we provide the first algorithm SCaLE...
Large language models generate judgments that resemble those of humans. Yet the extent to which these models align with human judgments in interpreting figurative and socially grounded language remains uncertain. To investigate this, human participants and four instruction-tuned LLMs of different sizes (GPT-4, Gemma-2-9B, Llama-3.2,...
End-to-end backpropagation couples all layers through a global error signal, enabling coordinated learning but requiring long-range credit assignment. Motivated by recent progress in blockwise self-supervised learning (BWSSL), we ask whether masked video transformers can be trained without end-to-end backpropagation. Applying BWSSL to...
Let a function $F: [0,1]^2\rightarrow [0,1]$ be given by $F(x,y)= f^{(-1)}(T(f(x), f(y)))$ where $f :[0,1]\rightarrow [0,1]$ is a monotone function, $f^{(-1)}$ is the pseudo-inverse of $f$ and $T$ is a triangular norm. This article characterizes the monotone function $f$ satisfying that the function $F$ is a conditionally cancellative...
Large language model (LLM) tokenizers act as structured compressors: by mapping text to discrete token sequences, they determine token count (and thus compute and context usage) and the statistical structure seen by downstream models. Despite their central role in LLM pipelines, the link between tokenization, compression efficiency...
We propose graph canonical coherence analysis (gCChA), a novel framework that extends canonical correlation analysis to multivariate graph signals in the graph frequency domain. The proposed method addresses challenges posed by the inherent features of graphs: discreteness, finiteness, and irregularity. It identifies pairs of...
Probabilistic computers built from p-bits offer a promising path for combinatorial optimization, but the dense connectivity required by real-world problems scales poorly in hardware. Here, we address this through graph sparsification with auxiliary copy variables and demonstrate two fully on-chip parallel tempering solvers on an FPGA....
Scientific reasoning increasingly requires linking structured experimental data with the unstructured literature that explains it, yet most large language model (LLM) assistants cannot reason jointly across these modalities. We introduce SpectraQuery, a hybrid natural-language query framework that integrates a relational Raman...
The parallel evolution of Large Language Models (LLMs) with advanced code-understanding capabilities and the increasing sophistication of malware presents a new frontier for cybersecurity research. This paper evaluates the efficacy of state-of-the-art LLMs in classifying executable code as either benign or malicious. We introduce an...
Prior work on LLM-based privacy focuses on norm judgment over synthetic vignettes, rather than how people think about a specific data practice and formulate their opinions. We address this gap by designing PrivacyReasoner, an agent architecture grounded in three key ideas: (1) LLMs can detect subtle privacy cues in natural language...
Large Language Models (LLMs) demonstrate potential to estimate the probability of uncertain events, by leveraging their extensive knowledge and reasoning capabilities. This ability can be applied to support intelligent decision-making across diverse fields, such as financial forecasting and preventive healthcare. However, directly...
Large Language Models (LLMs) motivate generative agent simulation (e.g., AI Town) to create a ``dynamic world'', holding immense value across entertainment and research. However, for non-experts, especially those without programming skills, it isn't easy to customize a visualizable environment by themselves. In this paper, we...
This study aims to conduct a bibliometric analysis of scientific publications discussing technology integration in the Project-Based Learning (PjBL) model during the 2015-2024 period. With the evolving 21st century educational paradigm, PjBL has emerged as one of the most promising pedagogical approaches. In this context, the...
This letter proposes a block sparse Bayesian learning (BSBL) algorithm of non-circular (NC) signals for direction-of-arrival (DOA) estimation, which is suitable for arbitrary unknown NC phases. The block sparse NC signal representation model is constructed through a permutation strategy, capturing the available intra-block structure...
Zero-Shot Anomaly Detection (ZSAD) leverages Vision-Language Models (VLMs) to enable supervision-free industrial inspection. However, existing ZSAD paradigms are constrained by single visual backbones, which struggle to balance global semantic generalization with fine-grained structural discriminability. To bridge this gap, we propose...
External memory systems are pivotal for enabling Large Language Model (LLM) agents to maintain persistent knowledge and perform long-horizon decision-making. Existing paradigms typically follow a two-stage process: computationally expensive memory construction (e.g., structuring data into graphs) followed by naive retrieval-augmented...
We present a transaction-driven dynamic reconfiguration protocol in Modern payment systems based on Byzantine Consistent Broadcast which can achieve high performance by avoiding global transaction ordering. We demonstrate the fundamental paradigm of modern payment systems, which combines user nonce based transactions ordering with...