ML Scaling

• The Practitioner’s Guide to the Maximal Update Parameterization | EleutherAI Blog

NLP

• [2203.15556] Training Compute-Optimal Large Language Models

• GitHub - huggingface/m4-logbook: experiments logbook

• GLM-130B/main-log-en.md at main · THUDM/GLM-130B · GitHub

CV

Multi Modal / Multi Task

• Open Flamingo
  • Knowledge sharing note - What did we learn in the past 3 months? - Google Docs
  • https://twitter.com/SanhEstPasMoi/status/1632775840135016448

Recommender Systems

• [2208.08489] Understanding Scaling Laws for Recommendation Models