Below is a list of some of my machine learning notes. I have only started sharing my notes publicly recently in the hope that they may be useful to others. I will continue to add notes to this page as I read new papers.

Language Models

Multi-Agent Motion Forecasting as Language Modeling

Understanding HTML with Large Language Models

LoRA

Prompt Tuning and Prefix Tuning

Extracting Training Data from Large Language Models

Byte Pair Encoding

Vision Models

BEVFusion

CenterPoint

DETR

Deformable DETR and Deformable Attention

Panoptic SegFormer

Swin and Swin v2

Concepts

Normalizing Flows

Differentially Private Training

Frameworks

Hugging Face Tokenizer

PyTorch Buffers