ML paper summaries

Summaries and personal thoughts of papers (mostly in machine learning) I have read in detail.

---

1. (Apr 18, 2026) Training Compute-Optimal Large Language Models
2. (Apr 18, 2026) Learning and Leveraging World Models in Visual Representation Learning
3. (Apr 2, 2026) FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness