Paper Search

Search a personal or lab library of papers by meaning.

Industry	Level	Difficulty	Recommended mode
Education & research	1 of 5	Beginner	flat

Coming soon

A complete, runnable Paper Search reference implementation is in progress. The architecture and core code below already work with SrvDB 0.3.

The problem

• Keywords miss papers using different terms.
• Libraries grow faster than memory.
• Notes are scattered.

How SrvDB powers Paper Search

1. Paper abstracts and paragraphs are embedded in SrvDB.
2. Questions find relevant passages.
3. Runs on a laptop.
4. New papers added with upsert.

What changes

Today	With Paper Search on SrvDB
Keyword search	Concept search
Forgotten papers	Rediscovered by meaning
Scattered notes	One searchable library

How it works

1. Extract text.
2. Embed paragraphs.
3. Search.
4. Save findings.
5. Add new papers.

SrvDB features it uses

• Exact search (flat): every query compares against every stored vector, so results are exact and explainable.
• upsert: inserts new items and replaces changed ones, so the knowledge base stays current.
• Crash-safe persist(): after it returns, data survives a crash or power loss and reopens exactly as committed.
• Fully offline: SrvDB runs in-process and never opens a network connection.

Use cases

• Literature search
• Related-work drafting
• Reading lists
• Citation hunting
• Thesis research
• Grant writing

Core code

embed() is your embedding model (for example sentence-transformers, or a function that turns your records into feature vectors).

import json
import srvdb

db = srvdb.Database("./paper-search-db", dimension=768)

def ingest(items):
    """items: list of (id, text, metadata dict)"""
    db.upsert(
        ids=[i for i, _, _ in items],
        embeddings=[embed(t) for _, t, _ in items],
        metadatas=[json.dumps(m) for _, _, m in items],
    )
    db.persist()                      # crash-safe checkpoint

def related_passages(question, k=5):
    hits = db.search(embed(question), k)
    return [(json.loads(db.get(i)), round(score, 3)) for i, score in hits]

Limits and honest notes

• PDF extraction quality matters.

Build roadmap

1. Prototype: load a sample of real data with flat mode and check that searches return what an expert expects.
2. Measure: if you move to an approximate mode, compare its results with flat on the same queries (recall).
3. Pipeline: add continuous upsert and decide when to call persist().
4. Harden: test restarts and crashes, add logging, review every output path with domain experts.
5. Deploy: ship the database directory with the application or service.

All Education & research projects

Level	Project	What it does
1 · Beginner	Paper Search (this page)	Search a personal or lab library of papers by meaning.
2 · Intermediate	Study Assistant	Answers questions only from the student’s own course material.
3 · Advanced	Reviewer Matcher	Matches papers or proposals to suitable reviewers.
4 · Expert	Research Group Brain	A shared, searchable memory for a lab: notes, data descriptions, figures.
5 · Frontier	Scientific Similarity Lab	Similarity search over spectra, molecules and other scientific data.

See also: Use cases · Live examples (coming soon)