Paper Notes

Notes on papers I read.

2025

October

Blockwise Parallel Transformer for Large Context Models | Paper Notes
AI Paper Notes

September

Fast Inference from Transformers via Speculative Decoding | Paper Notes
AI Paper Notes
Fast Transformer Decoding: One Write-Head is All You Need | Paper Notes
AI Paper Notes
GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints | Paper Notes
AI Paper Notes
Ring Attention with Blockwise Transformers for Near-Infinite Context | Paper Notes
AI Paper Notes
Effective Long-Context Scaling of Foundation Models | Paper Notes
AI Paper Notes
YaRN: Efficient Context Window Extension of Large Language Models | Paper Notes
AI Paper Notes

August

July