About
Research & Development
Beyond Relevance: How AI Models Are Learning to Pick Truly Useful Information for Better Answers
TLDR: This research introduces a method to distill the utility judgment capabilities of large language models (LLMs) into smaller, more efficient models for Retrieval-Augmented...
Research & Development
Securing Graph Data: A New Framework for Privacy-Preserving Structure Learning
TLDR: This research introduces a new framework for learning graph structures from open datasets while ensuring strong privacy guarantees using Gaussian Differential Privacy. Unlike...
Research & Development
Advancing Solar Panel Detection with Large Language Models
TLDR: The research introduces PVAL, a framework leveraging large language models (LLMs) for accurate and scalable solar panel detection, localization, and quantification from satellite...
Research & Development
AI’s Gender Problem: How Language Models Perpetuate Stereotypes in Italian
TLDR: A study investigated gender stereotypes in OpenAI ChatGPT and Google Gemini when responding to Italian prompts about professional roles. It found both models...
Research & Development
Precision in Diabetic Retinopathy: A Dual Attention Deep Learning Model
TLDR: A new deep learning model called CABNet, incorporating Global Attention Block (GAB) and Category Attention Block (CAB), significantly improves the accuracy of Diabetic...
Research & Development
Virne: A New Benchmark for Deep Reinforcement Learning in Network Resource Allocation
TLDR: Virne is a comprehensive benchmarking framework for evaluating deep Reinforcement Learning (RL) based methods in Network Function Virtualization Resource Allocation (NFV-RA). It offers...