https://schmoll.ai/posts/ Hands-on ML studies with real data and reproducible setups. en https://schmoll.ai/posts/foundation-models-head-ct/ https://schmoll.ai/posts/foundation-models-head-ct/ Fri, 18 Sep 2026 00:00:00 GMT A frozen, self-supervised foundation model — trained only on everyday photos, never on medical images — detects intracranial hemorrhage on head CT at ~0.94 AUC. What drives the transfer isn't what I bet on. https://schmoll.ai/posts/lejepa-text-pretraining/ https://schmoll.ai/posts/lejepa-text-pretraining/ Mon, 21 Sep 2026 00:00:00 GMT LeCun's newest self-supervised recipe trains video models at up to 20× less compute. I ported it to text and ran the controls. Result: a model that never reads more than 30% of any text learns better text representations than models that read everything — and the new invariance loss is not what makes that work.