Bernardo Ribeiro 👋

PhD Candidate at Eindhoven University of Technology, working on Multimodal AI for Anomaly Detection at the AIMS Lab. I hold a Cum Laude Master’s in Artificial Intelligence (2025) and my research interests include Generative Models, Resource-Efficient AI, and Computer Vision. Check out my research.

Recent Publications

View All →

FlowCast: Advancing Precipitation Nowcasting with Conditional Flow Matching
ICLR 2026
B. P. Ribeiro, J. F. Pucer
Developed the first end-to-end probabilistic model leveraging Conditional Flow Matching (CFM) for precipitation nowcasting. Achieved state-of-the-art accuracy on SEVIR/ARSO benchmarks while improving inference efficiency by 100x against Diffusion (DDIM) through optimized sampling (1–10 steps).
EdgeEar: Efficient Ear Recognition for Edge Devices
2025
C. Lendering*, B. P. Ribeiro*, et al.
Modified the EdgeNeXt CNN-Transformer architecture using Low-Rank Linear (LoRaLin) layers to achieve a 50x parameter reduction. Obtained the lowest Equal Error Rate (EER) on the UERC2023 dataset.

Recent Publications

FlowCast: Advancing Precipitation Nowcasting with Conditional Flow Matching

EdgeEar: Efficient Ear Recognition for Edge Devices