This project studies a promising alternative to current autoregressive (AR) large language models: diffusion-based large language models. It aims to make them generate high-quality text more efficiently while maintaining speed and energy efficiency. The research will focus on understanding how these models work, identifying their root limitations, and developing improved methods that balance text quality, […]

This project will study how well LoRaWAN can work for direct communication with Low Earth Orbit (LEO) satellites to connect IoT sensors in remote “no-coverage” areas. The intern will use a simulation tool to model thousands of devices sending data to a satellite during a single pass and will measure how often messages succeed or […]

Online hate is increasingly spread through memes, where harmful meaning comes from the combination of images and text, making it harder to detect than text-only abuse. Through a 12-week research visit to the University of Hildesheim, I will develop and evaluate efficient AI models that can better identify hateful memes across multiple datasets while remaining […]

This project seeks to improve the understanding of how Machine-Learning (ML)-specific code issues arise during software development by using Large Language Models to analyze developers’ code changes over time. An automated framework will be created to track code updates, detect ML-related issues, and identify the type of developer activity (feature development, bug fixing, enhancement, or […]

This project aims to build an AI system that can turn regular, human-written math problems and solutions into fully checked, error-free formal proofs, with a special focus on quantum mathematics. The project will collect open-source quantum math problems, train a machine-learning model to translate them into formal proofs using the Lean proof assistant, and create […]

Des microcommutateurs MEMS RF ont été fabriqués sur une ligne de prototypage industrielle de 200 mm. Compacts et à très faible consommation énergétique, ils sont particulièrement adaptés aux environnements cryogéniques où la dissipation thermique doit rester minimale. Le projet vise à qualifier ces dispositifs et des matrices de commutation à très basses températures dans un […]

L’émergence de l’informatique quantique marque une ère de transformation dans la science informatique, promettant une puissance de traitement sans précédent pour les problèmes complexes. Cependant, l’évolutivité de l’électronique de contrôle classique, à température ambiante, reste limitée pour les applications quantiques cryogéniques. L’équipe de Sherbrooke, en collaboration avec celle de Paris-Saclay, développe une plateforme électronique cryogénique […]

The project explores how to enable robots to perform multi-step tasks in realistic, changing environments by integrating visual information in different aspects of planning. Instead of assuming a perfect internal map of all objects, we will use camera input both to understand the scene and to guide the robot’s decisions during planning and execution. Demonstrations […]

Ce projet développe une plateforme neuromorphique innovante qui associe des circuits analogiques (CMOS et memristors) à des circuits numériques sur FPGA pour créer des réseaux de neurones inspirés du cerveau, capables d’apprendre de manière autonome. Cette technologie permettra des systèmes embarqués plus rapides, plus économes en énergie et plus fiables, adaptés à la robotique et […]

This Master’s thesis project, conducted under a Mitacs Globalink research internship, aims to address persistent challenges in hospital pharmaceutical supply management—such as fluctuating medication demand, supply-chain delays, and inefficient manual forecasting—by developing an end-to-end AI-enabled solution. Using a large dataset of 6 million prescription records and 2 million stock movement entries, the student will apply […]

In this project, we will design and develop an IoT device auto-detection model and deploy it on a Raspberry Pi. The model will follow a continuous machine-learning approach and will be trained using an IoT dataset developed in our lab. The Raspberry Pi will function as a Wi-Fi router, allowing IoT devices to connect to […]

Electric vehicle charging infrastructure has become a central component of transportation planning as adoption of electric vehicles accelerates worldwide. Modern research models how drivers choose a station using discrete choice models, especially multinomial logit, which capture realistic preferences such as distance, queues, and charging speed. These models are computationally expensive, making large-scale location planning challenging. […]