Quantifying the Pre-training Dividend: Generative versus Latent Self-Supervised Learning for Time Series Foundation Models

ArXi:2605.19462v1 Announce Type: cross The success of self-supervised learning (SSL) in vision and NLP has motivated its rapid adoption for time series. However, research has focused primarily on Generative paradigms and forecasting tasks, leaving the broader utility of learned representations unquantified. We establish a controlled framework to evaluate the "pre-