We do modern frequentist statistics: Using fake-data simulation

kept by eddie sep 18

The article explains how modern frequentist statistics can be practiced by simulating fake data to evaluate estimators, hypothesis tests, and confidence intervals under realistic conditions.

LAION Big Video Dataset
LAION releases a 10‑million‑hour open video dataset (LAION‑BVD) with 80M downloaded videos and synthetic captions for multimodal pre‑training across video, audio, and image modalities.
also on #data-extraction, #research, #machine-learning
Pretraining progress is mostly coming from data
From 2019 to 2025, data improvements contributed 3.24× more compute‑efficiency gains than model changes, delivering 12× versus 3.7× efficiency gains at 1e19 FLOPs.
also on #data-extraction, #machine-learning