awesome-embedding-models

Why we included this project

Text embeddings have changed a lot since word2vec and GloVe, and this list tracks that history. It gathers the foundational papers, the researchers behind them, and the courses and datasets used to train and evaluate the models, so you can trace how the field moved from static word vectors to BERT and sentence encoders. It is a reading list rather than something you deploy. The papers are grouped by approach, which makes it easy to jump to the level of representation you are working with. Anyone building NLP systems who wants to understand the trade-offs between embedding methods will find a useful map of where each idea came from and what it was meant to solve.