RAD-AI: Rethinking Architecture Documentation for AI-Augmented Ecosystems
This paper introduces RAD-AI, a framework that augments existing architecture documentation methods to better support the complexities of AI-augmented ecosystems, addressing compliance with the EU AI Act.
SAGAI-MID: A Generative AI-Driven Middleware for Dynamic Runtime Interoperability
SAGAI-MID is presented as a middleware solution utilizing large language models to dynamically resolve schema mismatches in modern distributed systems, enhancing interoperability.
EpiScreen: Early Epilepsy Detection from Electronic Health Records with Large Language Models
EpiScreen utilizes large language models on clinical notes to support early detection of epilepsy, enhancing diagnostic accuracy and reducing delays in patient care.
BIOGEN: Evidence-Grounded Multi-Agent Reasoning Framework for Transcriptomic Interpretation in Antimicrobial Resistance
BIOGEN is introduced as a multi-agent framework that enhances the interpretation of RNA-seq data in antimicrobial resistance research, promoting evidence-grounded analysis.
Trust-Aware Routing for Distributed Generative AI Inference at the Edge
This paper proposes G-TRAC, a trust-aware framework for ensuring robust distributed inference of generative AI across decentralized edge devices, enhancing reliability in real-time applications.
Fl-PBM: Pre-Training Backdoor Mitigation for Federated Learning
FL-PBM presents a proactive defense mechanism against backdoor attacks in federated learning environments, showcasing significant improvements in safeguarding model integrity.
Vision-Language Agents for Interactive Forest Change Analysis
An LLM-driven agent is introduced to facilitate forest change analysis through natural language querying, improving accessibility and interpretability of remote sensing data.
One stout to rule them all: Reconciling artificial intelligence, data science and malted alcoholic beverages
This study introduces a novel framework for analyzing craft beers using AI and data science, addressing challenges in understanding consumer trends in a growing market.
Can Hierarchical Cross-Modal Fusion Predict Human Perception of AI Dubbed Content?
A hierarchical framework is developed for evaluating AI-generated dubbed content through integrated analysis of audio, video, and text, enhancing automatic assessment capabilities.