RAG

Intermediate

Architecture that retrieves relevant documents (e.g., from a vector DB) and conditions generation on them to reduce hallucinations.

Full Definition

Architecture that retrieves relevant documents (e.g., from a vector DB) and conditions generation on them to reduce hallucinations.

Keywords

vector search grounding

Domains

Foundations & Theory

Related Terms

Semantic Search related to

Retrieval based on embedding similarity rather than keyword overlap, capturing paraphrases and related concepts.

 Chunking related to

Breaking documents into pieces for retrieval; chunk size/overlap strongly affect RAG quality.

 Grounding related to

Constraining outputs to retrieved or provided sources, often with citation, to improve factual reliability.

 SFT related to

Fine-tuning on (prompt, response) pairs to align a model with instruction-following behaviors.

 Autoregressive Model related to

Generates sequences one token at a time, conditioning on past tokens.

 Masked Language Model related to

Predicts masked tokens in a sequence, enabling bidirectional context; often used for embeddings rather than generation.

 Few-Shot Learning related to

Achieving task performance by providing a small number of examples inside the prompt without weight updates.

 Chain-of-Thought related to

Stepwise reasoning patterns that can improve multi-step tasks; often handled implicitly or summarized for safety/priv...

Concept Map

See how RAG connects to other concepts.

Open Knowledge Graph