# RAG Chunking and Retrieval Prep

Clean PDFs and Word files, remove extraction noise, scan hidden prompt risks, split text, score chunk quality, and package citation-ready RAG inputs.

> Canonical page: https://test.elysiatools.com/en/hubs/rag-chunking-retrieval-prep

- **Keywords:** rag chunking, retrieval prep, pdf to clean text, citation chunks

## Frequently asked questions

### Why not send the whole document directly to a vector database?

Raw documents often contain repeated headers, irrelevant pages, OCR noise, hidden text, and unstable boundaries. Cleaning and scoring first makes retrieval easier to debug and cite.

### When should I use sentence splitting instead of a generic text splitter?

Use sentence splitting when coherence matters and chunks should end at natural language boundaries. Use the generic text splitter for delimiter-based logs, transcripts, lists, or structured exports.

### Does this workflow create embeddings?

No. It prepares the corpus before embedding by producing cleaner text, chunk candidates, token estimates, quality scores, and citation metadata for the indexing step.

## Related content

- [PDF LLM and RAG Preparation](https://test.elysiatools.com/en/hubs/pdf-llm-rag-prep): Turn PDFs into clean, safe, citation-ready inputs for LLM summarization, embedding, search, and RAG pipelines.
- [Prompt Engineering and LLM Input Preparation](https://test.elysiatools.com/en/hubs/prompt-engineering-llm-input-workflows): Prepare LLM inputs by structuring prompts, checking token budgets, translating or identifying language, cleaning PDF evidence, and reviewing injection, math, regex, and data risks before use.
