Frequently Asked Questions

Yes. Domestic and international papers come back together in a single search.

• Japanese papers: Every search reaches papers from J-STAGE (about 4,500 Japanese journals publishing about 6 million articles) and other domestic journals through the JaLC (Japan Link Center) API.
• International papers: It also searches OpenAlex scholarly works from around the world (over 330M) in the same query.
• Ask in Japanese: The AI interprets your question and runs the domestic search alongside an English-language international search, then presents the results in Japanese.
• Toggle: You can choose whether to include domestic journals from a checkbox in the search form (on by default).

Note that items without DOIs, dissertations, and conference proceedings fall outside this coverage. For those, we recommend also checking CiNii.

Litra AI's AI is built to find and select papers, not to write text.

• Before the search: Even if your question is still vague, the AI asks follow-up questions to sharpen your topic and refine your search terms.
• During the search: Rather than skimming the web like general-purpose AI, it thoroughly collects 1,000+ papers from scholarly databases (OpenAlex, JaLC, J-STAGE), and the AI scores every single one against your request.
• After the search: Papers are ranked by score. When you open a paper, we check programmatically that the link works.

That is why it never suggests papers that don't exist, unlike general-purpose AI.

Litra AI does not ask an AI to write up papers. It finds them in databases of real papers.

• Collected from scholarly databases: It queries OpenAlex, JaLC, and J-STAGE directly and collects only papers that actually exist.
• No AI-made paper titles: Unlike ChatGPT, the AI does not generate text that names papers, so fabricated papers never slip in.
• Links checked when you open them: When you open a paper, we check programmatically that the link works, and look for another source if it doesn't.

It lets you read the full text of open-access papers with the original and its translation side by side (beta). Open it from "Analyze full text" on a paper card in your search results.

• Which papers: Open-access papers whose PDF is publicly available. Whether a translation is shown depends on the paper's license; papers that do not allow translation are shown in the original only.
• Papers per plan: The free tier includes 3 papers as a trial, Mini includes up to 3 papers a month, and Standard is unlimited.
• Credits: Each paper uses credits based on the actual AI cost. No credits are used if not a single page could be shown, for example when the PDF could not be retrieved. A paper you have read to the end stays in your browser on the same device for 30 days, so opening it again uses no credits.

Yes, absolutely. Litra AI is an academic paper search service, but it isn't limited to research use.

• Consultants: Use it to gather case studies as supporting evidence for client explanations.
• Businesspeople: Use it to back up proposals with citations.

Because papers carry more credibility than web articles, they're valued as solid evidence in external-facing materials.

Litra AI runs two kinds of search at once and merges the results.

• Keyword search: Like Google Scholar, it finds papers that match the terms you entered.
• Semantic search: It finds papers with similar meaning or concepts, even in different wording. For example, a search for 'crop monitoring' also finds papers about 'agricultural field observation' or 'farmland surveillance.'

Combining both means you don't miss relevant papers.

The papers we suggest are always real, but AI summaries and analysis are based on paper abstracts.

• Papers: Retrieved directly from the OpenAlex academic database, so only real papers are displayed.
• AI summaries, analysis, and research assistant: Generated by AI from paper abstracts. Useful for quickly grasping the big picture.

For deeper understanding, we recommend reading the original papers directly.

You can search scholarly papers from around the world and from Japanese journals.

• International papers (OpenAlex): 330M+ scholarly works. OpenAlex integrates over 50 data sources including arXiv and PubMed, covering 98.6% of PubMed across a wide range of disciplines.
• Japanese papers (JaLC): Litra AI also calls the JaLC (Japan Link Center) API at search time, so papers from J-STAGE and other domestic journals are included. You can toggle this from a checkbox in the search form (on by default).

Results typically arrive within 1-3 minutes.

• Progressive results: Results appear progressively during the search, so there's minimal perceived wait time.
• Compared to other services: Gemini Deep Research and ChatGPT Deep Research take 5-30 minutes; Litra AI delivers results significantly faster.

Litra AI works with abstracts and publicly available information only. We do not access paywalled full text.

• Paywalled papers: Use your institutional access or purchase them individually.
• Open-access papers: For arXiv, PubMed Central, and similar papers, we provide direct links to the full text.

The relevance score shows how well a paper matches your search, on a scale of 1 to 10.

• 8-10: Highly relevant, and specially highlighted in the search results.
• 5-7: Relevant. Check the content and judge for yourself.
• 1-4: Low relevance, worth only a quick glance.

The AI calculates the score from three factors: semantic match with your search, the abstract, and the title. The score does not indicate a paper's quality or importance, so a highly cited, well-known paper will still score low if it is far from your search topic.

Loosening your search conditions helps.

• Shorten your search: Packing in too many conditions narrows the candidates too far.
• Try other wording: If a technical term isn't landing, rephrasing it in more common words also helps (for example, "LLM" to "large language model").
• Answer the AI in detail: If you still find nothing, answering the AI's questions in more detail during the intent confirmation step improves accuracy.

Making your search more specific improves the scores.

• Describe your topic in detail: Instead of "machine learning", write something like "methods for improving accuracy in image classification using machine learning".
• Answer the AI's questions: Answering the AI's questions during the intent confirmation step improves accuracy further.
• Ask the research assistant: On the results screen, you can ask it to "find papers with higher relevance".

You can sort by three criteria. Use the buttons at the top of the search results screen to switch.

• Relevance (default): In order of the AI-calculated relevance score.
• Citations: By how often the papers have been cited by other papers.
• Year: Most recently published first.

When your search could be read in more than one way, we confirm which meaning you intend.

• Example: "Bias" could mean statistical bias or cognitive bias, and the papers you find will differ.
• How to answer: Just answer in your own words, and the AI will adjust the search based on your reply.
• In a hurry: You can skip the confirmation and start the search right away.

The free tier gives you 5 credits (approximately 5 searches). No credit card required.

• Features: You can try the same features as paid plans: multilingual translation, AI analysis chat, and paper analysis.
• AI analysis: AI analysis runs on Fast. Deep is available on paid plans, and Deep Pro on Standard and above.

Credits are the currency used for paper search and AI analysis on Litra AI.

• Per search: One search uses roughly 1 credit, from getting your results through a few messages with the AI.
• How usage is measured: Credits follow the actual AI usage cost, so the amount varies with what you do.
• Fast, Deep, and Deep Pro: Deep and Deep Pro think before answering, which suits detailed comparisons and synthesis. Deeper tiers think more and use slightly more credits.

Choose by how often you search. One search, including the follow-up conversation with the AI, uses about 1 credit.

• A few times a week (up to about 30 times/month): Mini ($9/mo, 30 credits) is recommended. Great for light literature searches and report writing.
• About 3 times a day (up to about 100 times/month): Standard ($19/mo, 100 credits) is recommended. Ideal for regular paper checks and research activities. In months when you need more, you can buy extra one-time credits.

Mini and Standard differ in credits (30 vs 100), login devices (2 vs 3), and full-text paper analysis (3 papers a month vs unlimited).

Yes, unused credits roll over to the next month.

• Carryover cap: You can hold up to 2x your monthly credits. For example, on Standard (100 credits/month), you can hold up to 200 credits including carried-over amounts.
• One-time credits: They are not subject to this cap and stay available for 12 months from purchase.

We accept credit cards (Visa, Mastercard, American Express, JCB).

All transactions are securely processed through Stripe.

Yes. We support reimbursement-based payment for KAKENHI.

• Receipt: After paying by credit card, a receipt is issued automatically. You can submit it to your university's accounting department.
• Company seal: The receipt does not include a company seal (inkan), but this is legally fine. If your university requires a seal, please contact us and we'll accommodate.
• Invoiced payment: Bank transfer by invoice is planned for the future.

Yes, you can cancel at any time.

After cancellation, you can continue using the service until the end of your current billing period.

If you purchased an annual plan within the last 14 days, you can request a refund from your account page.

• Refund amount: Prorated based on unused credits (refund = payment × remaining credits ÷ total credits).
• Processing fee: A processing fee (3.6%) may be deducted from refunds.
• Monthly plans: Monthly plans do not offer prorated refunds, but you can continue using the service until the end of the billing period.

We do not offer an enterprise plan or a plan for research labs.

For lab use, each member can subscribe to Standard ($19/mo, 100 credits) and buy extra one-time credits in months when they need more.

Getting your search results and the AI's first analysis uses roughly 1 credit.

• Why it varies: Credits follow the actual AI usage cost. The AI model used for search is fixed on our side, so this is the same on every plan.
• If your balance drops faster: Continued use of the research assistant (AI chat) on the results screen may be included. Each chat message uses only a small fraction of a credit, slightly more on deeper tiers (Deep and Deep Pro).

Google Scholar is an excellent search engine, but it only supports keyword search. Litra AI takes you from search to analysis in one place.

• Search: Combining keyword and semantic search finds related papers even when they use different terminology.
• Visualization: Results are shown in a tree map, so you can see how papers relate at a glance.
• Narrowing down: An interactive research assistant helps you narrow down to exactly the papers you need, with no need to search Google Scholar and then upload to NotebookLM.

Detailed comparison with Google Scholar

Elicit and SciSpace are excellent AI paper search services, but their UIs are English-only and require you to craft your own search queries.

• Ask in your language: Just describe what you need in your own language, and receive organized results in report format.
• Refine by dialogue: Talking with the AI lets you narrow down to exactly the papers you need.

It is ideal for researchers who find searching itself challenging.

Detailed comparison with Elicit

Consensus specializes in Yes/No research questions and supports English only.

• Where Consensus fits: Either-or questions like 'Does Vitamin D reduce cancer risk?'
• Where Litra AI fits: Open-ended, exploratory research like 'Survey the latest crop monitoring methods,' in your own language.

Detailed comparison with Consensus

Try the steps for your situation below.

• Forgot your password: Reset it from "Forgot your password?" on the sign-in page.
• Reset email not arriving: Check your spam folder and confirm you entered the same email address you registered with. Delivery can take a few minutes.
• Registered with Google: No password reset is needed. Sign in with the "Sign in with Google" button.
• Google sign-in error: Try clearing your browser cache and cookies, disabling your popup blocker, and then a different browser, in that order.