large language models

Beyond Mueller–Hinton Broth: A Scoping Review of Metabolic Context, Conditional Antibiotic Susceptibility, and Metabolism-Aware Antimicrobial Susceptibility Testing

Background. Standard antimicrobial susceptibility testing (AST) in Mueller–Hinton broth samples a single metabolic state, yet bacterial susceptibility in vivo depends on nutrient availability, host-derived metabolites, and spatial structure. The extent to which this mismatch produces clinically relevant errors in susce...

Token-Generation-Control-Deterministic-Logical-Timeline

Text generation by large language models is essentially a probabilistic sampling process: the model computes a probability distribution over the next token based on the input context, and then samples a token from that distribution. This probabilistic design gives large models creativity but also introduces a fundament...

A Lightweight Hybrid Framework for Detecting AI-Generated Scientific Text Using Multi-granularity Features and Language-Model Surprisal

The increasing adoption of generative artificial intelligence (Gen-AI) in scientific writing has transformed scholarly communication while also raising concerns about authorship transparency and research integrity. The growing use of large language models (LLMs) has increased the need for methods that can screen AI-gen...

PEDAGOGICAL CONDITIONS FOR USING ARTIFICIAL INTELLIGENCE TECHNOLOGIES IN PREPARING FUTURE TEACHERS FOR PROFESSIONAL ACTIVITY

This thesis investigates the structured integration of artificial intelligence (AI) tools within higher pedagogical education under specialty 13.00.01. As educational paradigms shift globally towards digital environments, future educators must develop comprehensive digital-pedagogical competencies. The study highlights...

Prioritising substances of concern for monitoring by using knowledge graphs and public datasets

Abstract Air pollution is identified by the World Health Organisation (WHO) as the leading environmental threat to human health, with indoor exposure particularly concerning given its daily, long-term nature. Europeans spend approximately 90% of their time indoors, where poor indoor air quality (IAQ) contributes an est...

Data and Code of Solution-Focused Brief Therapy Training with a Large Language Model-Based Virtual Standardized Patient

Dataset Systematic Literature Review: AI Generatif, Deskilling, dan Pengawasan Editorial dalam Jurnalisme

Review: AI Generatif, Deskilling, dan Pengawasan Editorial dalam Jurnalisme Pembuat: Novanka, Magister Ilmu Komunikasi, Fakultas Ilmu Sosial dan Ilmu Politik, Universitas Pelita Harapan (novankawijaya@gmail.com) Delliana Fitriah, Magister Ilmu Komunikasi, Fakultas Ilmu Sosial dan Ilmu Politik, Univeristas Pelita Harapa...

ClaimArena-LLM: Auditing Adversarial Propagation of False Causal Attributions in Multi-Agent LLM Debates

Background: Multi-agent large language model (LLM) systems increasingly rely on structured interaction, debate, and final model-based evaluation, yet an unsupported causal attribution introduced by one participant may propagate through the interaction and influence subsequent decisions. This study introduces ClaimArena...

RoboHarness: A Simple Harness Outperforms VLA and World Action Models

The rise of large language models (LLMs) has opened a path toward generalist embodied AI, raising the question of how LLMs can act in the physical world. One straightforward approach is training the model to generate actions: collect robot-action data and video demonstrations, refine the LLM architecture with an action...

Sustainable and Synergistic Energy Management for Large Model Computing

The rapid growth of foundation models, most prominently large language models (LLMs), is turning the electricity consumption of computing infrastructure into a first-order constraint on the artificial intelligence (AI) industry. This review synthesizes recent literature on three levers for reconciling large-model compu...

Vulnerabilities of LEO Megaconstellations and Cascading Threats to Civilian Populations: Dual-Use Militarization, RF Signature Geolocation, and Systematic Infrastructure Exposure

Markdown 【English Abstract】Low Earth Orbit (LEO) satellite mega-constellations, exemplified by SpaceX's Starlink, have emerged as critical planetary communication infrastructure. However, the accelerating integration of civilian networks with dedicated defense architectures such as "Starshield" profoundly destabilize...

Semantic Privilege Collapse: Trust Boundaries and Execution-Level Authorization in Agentic AI Systems

Large language model agents operating within tool-enabled pipelines create opportunities for untrusted content to influence consequential actions. This note introduces semantic privilege collapse as an architectural failure condition in which task data or insufficiently authorized instructions are promoted into effecti...