Masked Language Modeling

Domain: NLP Fundamentals

🎯 Motivation

Hides some tokens and asks the model to predict them. How BERT is trained.

📋 Concrete Example

The [MASK] sat on the mat. The model learns to predict cat from the surrounding context.

🔗 Analogy

Like a fill-in-the-blank quiz - the model learns language by guessing the hidden words.