RC RANDOM CHAOS

representation learning

2 posts

NovaMind ditches SSL loss for probe accuracy
Article

NovaMind ditches SSL loss for probe accuracy

Why pretext loss misleads in self-supervised learning, and how to select hyperparameters and architectures using probing harnesses that actually track quality.

 Sub-JEPA tightens the prediction signal
Article

Sub-JEPA tightens the prediction signal

Sub-JEPA is a small loss-side fix to LeCun's world models that consistently improves performance. Here's how it works, where it fails, and why it matters.