VAMP-Diff: VampPrior Latent Diffusion for Photoplethysmography Modeling

This collection of recent arXiv submissions (eess_sp) presents applied and methodological advances across biomedical signal modeling, wireless/radar channel representation, robust learning, and datasets. Key contributions include: VAMP-Diff — a VampPrior-regularized latent diffusion model for realistic photoplethysmography (PPG) generation and reconstruction that preserves heart/respiratory physiology; a tutorial and methods overview for Topological Signal Processing (TSP) on simplicial complexes; PilotWiMAE — a pilot-native self-supervised encoder for wireless channels with released pretrain/