AI Encyclopedia · AIPedia

Foundations

LLM (Large Language Model): How They Work and What They Are Good For

A large language model is a transformer-based neural network trained on massive text corpora to predict the next token. The mechanism, what it can and cannot do, and how to choose the right model.

Techniques

RAG (Retrieval-Augmented Generation): How LLMs Use External Knowledge

Retrieval-Augmented Generation lets large language models cite up-to-date sources by retrieving relevant documents before generating answers. The architecture, trade-offs, and when to use RAG over fine-tuning.