Short answer
Use normal search when you know the words. Use Semantic Search BETA when you know the idea but not the wording. Install one local model, build the project index, and open every suggested passage before you use it.
This guide covers CiteFlow. Check requirements and download options before you begin.
On this page
When to use this method
Use this method when different papers may describe the same idea with different words. It searches PDFs already in your project.
Before you start
Add the PDFs. Apply OCR to image-only pages. Leave enough disk space for one model and its index. Exact search finds matching words. Semantic Search BETA finds related meaning. An index is a local map of the project text used for later searches.
Technical note Confirmed behavior and an important limit.
- What we confirmed
- PDF extraction, embeddings and queries run on the PC after the optional model download. Embeddings are numerical representations used to compare meaning. Similarity scores identify related passages but do not assess truth, study quality or applicability.
- Important limit
- A high-ranking passage about the same intervention can use a different population. Read the surrounding method and limitation before treating it as evidence for the review question.
Match the query to what you know
Search exactly for a known phrase such as sleep duration or a source identifier. Try a conceptual query such as how shorter sleep affects attention when papers may use different wording. These are illustrative queries, not validated search strategies. Open the surrounding passage to check who was studied, what was measured and what the authors actually concluded.
Install, index and verify a local model
-
Try exact search first
Search for a known phrase, title, or identifier. You should see direct word matches without installing a semantic model.
-
Choose one model
E5 Large INT8 is the smaller download and uses less memory. BGE-M3 is the larger multilingual option. Choose one that fits your computer, then test it against passages you already know; a larger model is not a guarantee of better matches for your material.
-
Install and build the index
Let CiteFlow verify the model and index the project text. You should see the component and index marked ready before searching.
-
Search for one clear idea
Choose the source scope and enter one specific idea. You should see passages with source names and pages that you can open.
-
Try an unrelated control query
A control query is a test phrase unrelated to your topic. The search may still return the closest available passages. A high rank only compares results within that query; it is not a confidence threshold. Reject unrelated passages. Inspect source text and index state when a known relevant query misses an expected passage.
-
Fix an incomplete index
If the component is missing or failed, use Manage to install or repair it. Apply OCR when scanned pages are marked partial. Rebuild a stale index after source changes. Restart an index that was canceled.
-
Open and check each result
Open a promising result. You should see the original page and surrounding text. Check its population, method, qualifiers, and source before keeping or exporting it.
Model size and memory details
E5 Large INT8 is an approximately 557 MiB download and needs at least 4 GiB of available RAM; 8 GiB of system RAM is recommended. BGE-M3 Full Precision is an approximately 2.13 GiB download and needs at least 6 GiB available; 16 GiB of system RAM is recommended. Available RAM is the memory left for the task while Windows and other apps are running. Leave additional disk space for the index. Each model builds its own local project index.
Semantic-search limits
- Installing a large model for an exact identifier search
- Treating rank as study quality
- Ignoring OCR or extraction errors in indexed text
- Citing a match without reopening its page
Checkpoint: Retrieval checks
- Compare one normal text query with one conceptual query.
- Confirm the computer has enough storage and memory for the selected model.
- Run an unrelated test phrase and reject irrelevant results regardless of their rank.
- Open the original page for each retained match.
- Confirm exported matches use the intended source scope.