Literature Search Basics

DeepPaper AI makes it easy to find academic papers. This guide shows you how to use the search feature.

Search Interface

To search for papers:

1. Click "Literature Search" in the navigation bar
2. You'll see the search form
3. Type your keywords and pick a search type

Note: Search only works with English keywords. Use English for best results.

Starting a Search

Basic Search

1. Go to the search page
2. Type your English keywords (like "machine learning" or "neural network")
3. Pick a search type
4. Click "Search"

Search Types

You can search for different kinds of papers:

• General: Regular paper search
• Survey: Review papers
• Benchmark: Papers with performance comparisons
• Tutorial: How-to papers
• Perspective: Opinion pieces
• Dataset: Papers about datasets

What Happens Next

After you search, the system will:

1. Show the search status
2. Find relevant papers
3. List the results
4. Create a summary

Search Status

You'll see one of these statuses:

• Pending: Search is waiting to start
• Processing: Search is running
• Completed: Results are ready
• Failed: Something went wrong

Search History

You can see all your past searches:

1. Go to the search page to see your history
2. Each search shows keywords, type, status, and when you ran it
3. Click any search to see its results
4. Delete searches you don't need anymore

Common Questions

Why is my search taking so long?

Searching connects to academic databases and processes lots of data. It might take longer during busy times. If it's stuck, try refreshing the page or coming back later.

What if my search fails?

If you see "Failed", it could be because:

• Network problems
• Service is down
• Wrong keyword format

Try checking the error message, changing your search, or trying again later.

Can I run multiple searches?

Yes, you can start several searches. The system will handle them one at a time. Check your search history to see all their statuses.

Where do the papers come from?

Right now, DeepPaper AI gets papers from arXiv. We'll add more sources in future updates.

Why English only?

The search uses arXiv, which mostly has English papers with English metadata. Using English keywords gives you the best results.