Retrieval-Augmented Generation Cheat Sheets

Agentic RAG cheat sheet

An agent loop only pays off on top of retrieval that already works, and only when it has a hard stopping rule.

#Retrieval-Augmented GenerationAnimated

→ Topic: Agentic RAG

Contextual Retrieval cheat sheet

Contextual retrieval pays once at indexing time so every chunk carries the meaning it lost when the document was split.

#Retrieval-Augmented GenerationAnimated

→ Topic: Contextual Retrieval

Embedding cheat sheet

Distance carries meaning only inside one model's space, so the index and every query must share that model.

#Retrieval-Augmented GenerationAnimated

→ Topic: Embedding

Long-Context vs RAG cheat sheet

Long context lets attention sort the whole corpus; RAG filters before the prompt, so each fails in its own way.

#Retrieval-Augmented GenerationAnimated

→ Topic: Long-Context vs RAG

Multi-Vector Retrieval cheat sheet

Keeping one vector per token recovers nuance a single embedding averages away, but the index grows accordingly.

#Retrieval-Augmented GenerationAnimated

→ Topic: Multi-Vector Retrieval

Query Transformation cheat sheet

Each rewrite technique fixes one kind of retrieval failure, and every rewrite adds a call that can drift or invent.

#Retrieval-Augmented GenerationAnimated

→ Topic: Query Transformation

RAG Evaluation cheat sheet

A wrong answer has two possible causes, so a faithful answer built on the wrong chunks still needs a retrieval fix.

#Retrieval-Augmented GenerationAnimated

→ Topic: RAG Evaluation

RAG Guardrails and Grounding cheat sheet

A guardrail catches an unsupported claim before the user reads it, but it does not make the model itself more accurate.

#Retrieval-Augmented GenerationAnimated

→ Topic: RAG Guardrails and Grounding

Reranking cheat sheet

A reranker sharpens precision on what retrieval already found; it cannot rescue a document that never made the list.

#Retrieval-Augmented GenerationAnimated

→ Topic: Reranking

Retrieval-Augmented Generation cheat sheet

Good retrieval is necessary, not sufficient: generation can still ignore the right document.

#Retrieval-Augmented GenerationAnimated

→ Topic: Retrieval-Augmented Generation

Sentence Transformers cheat sheet

Contrastive training, not the base model, is what makes sentence vectors comparable by meaning.

#Retrieval-Augmented GenerationAnimated

→ Topic: Sentence Transformers

Similarity Search Algorithms cheat sheet

Approximate search gives up a little accuracy for speed; weak results usually start in the embeddings, not the index.

#Retrieval-Augmented GenerationAnimated

→ Topic: Similarity Search Algorithms

Sparse Retrieval cheat sheet

BM25 and learned sparse models share one index and its limits; they differ only in who decides the weights.

#Retrieval-Augmented GenerationAnimated

→ Topic: Sparse Retrieval

Vector Indexing cheat sheet

An index trades a little recall for speed; tune it against an exact baseline so you know what it skips.

#Retrieval-Augmented GenerationAnimated

→ Topic: Vector Indexing