About

I’m a second-year MSc student at UFMG, Brazil, supervised by Wagner Meira Jr. working on self-supervised learning and world models. I did my undergrad in electrical engineering there, then spent two years in a double degree program at Télécom Paris (Institut Polytechnique de Paris), where I also earned a French engineering degree (Diplôme d’Ingénieur).

I also collaborate with Randall Balestriero on his stable family repositories (worldmodel, datasets, pretraining), JEPA architectures, and world models more broadly. I believe research should be open source, and that SSL and reward-free methods are the right path toward better generalization and transfer for world models.

Outside of work, I’m really into cinema — French Nouvelle Vague and Japanese film in particular — and literature, especially art criticism and essays (Sontag, Barthes, etc.). If you want to talk movies, find me on Letterboxd. Beyond that, I like cooking, photography, traveling, and follow volleyball (although I don’t play). I’ve also been reading One Piece weekly since 2013, so it’s been a long journey.

Publications

1. stable-worldmodel: A Platform for Reproducible World Modeling Research and Evaluation
   Lucas Maes, Quentin Le Lidec, Luiz Facury, Nassim Massaudi, Ayush Chaurasia, Francesco Capuano, Richard Gao, Taj Gillin, Dan Haramati, Damien Scieur, Yann LeCun, Randall Balestriero
   ArXiv, 2026

2. Learning Navigable World Models via Latent Energy Shaping
   Luiz Facury, Jose Fernandes, Pedro Dutenhefner, Gisele Pappa, Wagner Meira Jr.
   2nd Workshop on World Models @ ICLR 2026

3. Beyond Patient Invariance: Learning Cardiac Dynamics via Action-Conditioned JEPAs
   Jose Fernandes, Luiz Facury, Pedro Dutenhefner, Gisele Pappa, Wagner Meira Jr.
   2nd Workshop on World Models @ ICLR 2026

4. Transferring Clinical Knowledge into ECGs Representation: A Self-Supervised Approach for Interpretable, Unimodal-at-Inference Diagnosis
   Jose Fernandes, Luiz Facury de Souza, Pedro Dutenhefner, Gisele Pappa, Wagner Meira Jr.
   TS4H @ NeurIPS 2025

5. Curved Spaces, Enhanced Diagnosis: Hyperbolic Neural Networks for Multi-label ECG Classification
   Pedro Dutenhefner, Diogo Tuler, Turi Rezende, José Fernandes, Luiz Facury, Luísa Porfírio, Yan Aquino, Arthur Buzelin, Pedro Bento, Gabriela Piaxão, Gisele Pappa, Antonio Ribeiro Wagner Jr.
   Computing in Cardiology (CinC), 2025

6. Clinically Interpretable Zero-Shot ECG Classification via Multimodal Learning and Expert-Aligned Descriptors
   Luiz Facury de Souza, Jose Fernandes, Pedro Dutenhefner, Turi Rezende, Gisele Pappa, Gabriela Paixão, Antonio Ribeiro Wagner Jr.
   CINC 2025/LXAI @ ICML 2025 (Selected for oral presentation at LXAI @ ICML)

7. Dense Self-Supervised Learning for Medical Image Segmentation
   Maxime Seince, Loic Le Folgoc, Luiz Augusto Facury de Souza, Elsa Angelini
   Medical Imaging with Deep Learning (MIDL), 2024

8. Multiple-Input–Multiple-Output Randomized Fuzzy Cognitive Map Method for High-Dimensional Time Series Forecasting
   Omid Orang, Hugo Vinicius Bitencourt, Luiz Augusto Facury de Souza, Patrícia de Oliveira Lucas, Petrônio CL Silva, Frederico Gadelha Guimarães
   IEEE Transactions on Fuzzy Systems, 2024

9. Combining embeddings and fuzzy time series for high-dimensional time series forecasting in internet of energy applications
   Hugo Vinicius Bitencourt, Luiz Augusto Facury de Souza, Matheus Cascalho dos Santos, Rodrigo Silva, Petrônio Cândido de Lima e Silva, Frederico Gadelha Guimarães
   Elsevier Energy, 2023

10. An embedding-based non-stationary fuzzy time series method for multiple output high-dimensional multivariate time series forecasting in IoT applications
    Hugo Vinicius Bitencourt, Omid Orang, Luiz Augusto Facury de Souza, Petronio CL Silva, Frederico Gadelha Guimaraes
    Springer Neural Computing and Applications, 2023