Podcast

Mechanical Dreams

An automatically generated podcast about machine learning and natural language processing. The two fictional hosts talk about papers that I want to learn more about on my way to work. It's not good,…

Download on the App Store

Already have it? Open in the app

Latest episodes

  1. QuaRot8 Aug 2026
  2. How to set AdamW’s weight decay as you scale model and dataset size7 Aug 2026
  3. Weight Decay Improves Language Model Plasticity7 Aug 2026
  4. Bridging Compute- and Data-Optimal Pretraining30 Jul 2026
  5. Pushing LLM Pretraining Scales29 Jul 2026
  6. Why Larger Models Learn More16 Jul 2026
  7. Rethinking the Role of Efficient Attention in Hybrid Architectures15 Jul 2026
  8. Nemotron-Labs-Diffusion14 Jul 2026
  9. Lightning Index Ultra-Long Context via Lookahead Sparse Attention13 Jul 2026
  10. Learning beyond Teacher12 Jul 2026