naacl_transfer_learning_tutorial

Why we included this project

This repository holds the code that accompanied the NAACL 2019 tutorial on transfer learning in NLP, taught by Sebastian Ruder, Matthew Peters, Swabha Swayamdipta, and Thomas Wolf. It walks through the main transfer learning techniques in their simplest form, deliberately favoring clarity over state-of-the-art performance, so you can trace each method end to end instead of digging through a large production codebase. The examples pair with the tutorial slides and a Google Colab notebook, which makes running them easy without a local setup. If you are trying to understand how modern pretrained language models work and how to adapt them to downstream tasks, this is a compact starting point, and it stays useful as an educational reference even years after the conference.