arXiv preprints from January 1, 2026 through July 28, 2026 — 10:20:49 EST
The Transformer architecture, a cornerstone of modern Large Language Models (LLMs), has achieved extraordinary success in sequence modeling, primarily due to its attention mechanism. However, despite its power, the standard attention mechanism is plagued by well-documented issues: representational collapse and attention sink. Although...
Question-answering (QA) models have advanced significantly in machine reading comprehension but often exhibit biases that hinder their performance, particularly with complex queries in adversarial conditions. This study evaluates the ELECTRA-small model on the Stanford Question Answering Dataset (SQuAD) v1.1 and adversarial datasets...
While foundation models in radiology are expected to be applied to various clinical tasks, computational cost constraints remain a major challenge when training on 3D-CT volumetric data. In this study, we propose TotalFM, a radiological foundation model that efficiently learns the correspondence between 3D-CT images and linguistic...
The rapid advancement of Large Language Models (LLMs) presents new opportunities for automated software vulnerability detection, a crucial task in securing modern codebases. This paper presents a comparative study on the effectiveness of LLM-based techniques for detecting software vulnerabilities. The study evaluates three approaches,...
Android Permission Model and Application (app) analysis has consistently remained the focus of the investigation of research groups and stakeholders of the Android ecosystem since it was launched in 2008. Even though the Android smartphone operating system (OS) permission model has evolved significantly from `all-or-none access' to...
As wireless networks continue to evolve, stringent latency and reliability requirements and highly dynamic channels expose fundamental limitations of gNB-centric massive multiple-input multiple-output (mMIMO) architectures, motivating a rethinking of the user equipment (UE) role. In response, the UE is transitioning from a passive...
Accurate and efficient classification of Alzheimer's disease (AD) severity from brain magnetic resonance imaging (MRI) remains a critical challenge, particularly when limited data and model interpretability are of concern. In this work, we propose TDA-Alz, a novel framework for four-stage Alzheimer's disease severity classification...
The rapid growth of artificial intelligence (AI) has brought novel data processing and generative capabilities but also escalating energy requirements. This challenge motivates renewed interest in neuromorphic computing principles, which promise brain-like efficiency through discrete and sparse activations, recurrent dynamics, and...
Effective pest management is crucial for enhancing agricultural productivity, especially for crops such as sugarcane and wheat that are highly vulnerable to pest infestations. Traditional pest management methods depend heavily on manual field inspections and the use of chemical pesticides. These approaches are often costly,...
This paper reveals that LLM-powered agents exhibit not only demographic bias (e.g., gender, religion) but also intergroup bias under minimal "us" versus "them" cues. When such group boundaries align with the agent-human divide, a new bias risk emerges: agents may treat other AI agents as the ingroup and humans as the outgroup. To...
We propose the Consensus-Based Privacy-Preserving Data Distribution (CPPDD) framework, a lightweight and post-setup autonomous protocol for secure multi-client data aggregation. The framework enforces unanimous-release confidentiality through a dual-layer protection mechanism that combines per-client affine masking with...
Transformer residual streams evolve by additive accumulation: each layer appends a feature update to a shared hidden state, but has no direct mechanism for replacing content that has become obsolete or conflicting. We introduce Deep Delta Learning (DDL), a residual update rule that preserves the identity path while giving every layer...
Functional connectivity (FC) analysis, a valuable tool for computer-aided brain disorder diagnosis, traditionally relies on atlas-based parcellation. However, issues relating to selection bias and a lack of regard for subject specificity can arise as a result of such parcellations. Addressing this, we propose ABFR-KAN, a...
Adapting to unforeseen novelties in open-world environments remains a major challenge for autonomous systems. While hybrid planning and reinforcement learning (RL) approaches show promise, they often suffer from sample inefficiency, slow adaptation, and catastrophic forgetting. We present a neuro-symbolic framework integrating...
We present judgeWEL, a dataset for named entity recognition (NER) in Luxembourgish, automatically labelled and subsequently verified using large language models (LLM) in a novel pipeline. Building datasets for under-represented languages remains one of the major bottlenecks in natural language processing, where the scarcity of...
Log data plays a critical role in observability, debugging, and performance monitoring in modern cloud-native systems. In small and early-stage cloud deployments, however, log retention policies are frequently configured far beyond operational requirements, often defaulting to 90 days or more, without explicit consideration of their...
This thesis investigates two key phenomena in large language models (LLMs): in-context learning (ICL) and model collapse. We study ICL in a linear transformer with tied weights trained on linear regression tasks, and show that minimising the in-context loss leads to a phase transition in the learned parameters. Above a critical...
Detecting coordinated inauthentic behavior on social media remains a critical and persistent challenge, as most existing approaches rely on superficial correlation analysis, employ static parameter settings, and demand extensive and labor-intensive manual annotation. To address these limitations systematically, we propose the Adaptive...
Current resource allocation paradigms, particularly in academic evaluation, are constrained by inherent limitations such as the Matthew Effect, reward hacking driven by Goodhart's Law, and the trade-off between efficiency and fairness. To address these challenges, this paper proposes "Bit-politeia", an AI agent community on blockchain...
Roadside litter poses environmental, safety and economic challenges, yet current monitoring relies on labour-intensive surveys and public reporting, providing limited spatial coverage. Existing vision datasets for litter detection focus on street-level still images, aerial scenes or aquatic environments, and do not reflect the unique...
Deploying LLMs efficiently requires testing hundreds of serving configurations, but evaluating each one on a GPU cluster takes hours and costs thousands of dollars. Discrete-event simulators are faster and cheaper, but they require re-implementing the serving system's control logic -- a burden that compounds as frameworks evolve. We...
Prosecutors across Mexico face growing backlogs due to high caseloads and limited institutional capacity. This paper presents a machine learning (ML) system co-developed with the Zacatecas State Prosecutor's Office to support internal case triage. Focusing on the Módulo de Atención Temprana (MAT) -- the unit responsible for intake and...
In this paper, we propose NeoVerse, a versatile 4D world model that is capable of 4D reconstruction, novel-trajectory video generation, and rich downstream applications. We first identify a common limitation of scalability in current 4D world modeling methods, caused either by expensive and specialized multi-view 4D data or by...
$\def\DD{\bf δ}\def\CH{\mathop{\mathrm{ConvexHull}}}\newcommand{\LL}{\cal {L}} \newcommand{\ZZ}{\mathbb{Z}} $ Given a convex body $C$ in the plane, its discrete hull is $C^0 = \CH( C \cap \LL )$, where $\LL = \ZZ \times \ZZ$ is the integer lattice. We present an $O( |C^0| \log \DD(C) )$-time algorithm for calculating the discrete hull...
Human detection in videos plays an important role in various real-life applications. Most traditional approaches depend on utilizing handcrafted features, which are problem-dependent and optimal for specific tasks. Moreover, they are highly susceptible to dynamical events such as illumination changes, camera jitter, and variations in...