---
title: "Search academic papers with Scholar"
description: "Ask a research question and get an answer built on papers, with each claim cited to a passage you can open and check."
canonical: "https://bearly.ai/docs/features/scholar/"
markdown: "https://bearly.ai/docs/features/scholar.md"
---

# Search academic papers with Scholar

HTML: [https://bearly.ai/docs/features/scholar/](https://bearly.ai/docs/features/scholar/)

> Ask a research question and get an answer built on papers, with each claim cited to a passage you can open and check.

Scholar gives a chat a paper index beside Web Search. When a question should rest on research literature, Bearly searches papers, reads the passages that carry a claim, and cites them with superscript numbers. A **Papers** card under the reply lists what it found.

Scholar works in Bearly on the web, desktop, iOS, and Android, on every plan. A team's policy can turn it off.

## Ask a research question

Ask in ordinary language, the way you would put the question to a colleague in the field. Scholar is a better fit than Web Search for scientific, medical, and technical questions, and for anything that asks what the research says.

```text copy
What does the research say about whether sleep deprivation impairs working memory in adults?
```

Bearly decides when papers are the right source. During the reply you see **Searching papers**, then **Found 8 papers**, and often **Reading paper** as it pulls passages from the papers that carry the load-bearing claims. News, products, companies, and how-to questions still go to Web Search.

Citations appear as superscript numbers. Hover a number, or press and hold on a touch screen, to see the paper: where it lives, its title, authors, year, and the opening of its abstract. Select the number to open the paper.

## Read the Papers card

The **Papers** card under a reply names what was searched and how many papers came back. Expand it to see each paper with its source, authors, year, and field. Select a paper to open it.

The source tells you what kind of document you are looking at:

- **arXiv**, **bioRxiv**, and **medRxiv** are preprint servers. Bearly says so when a finding rests only on a preprint.
- **PubMed** and **PMC** are the biomedical literature.
- A website name, such as a journal's site, means Bearly found the paper through an academic web search rather than in the paper index. That search uses Web Search, so it needs Web Search available in the chat.

When Bearly reads a paper, the card marks it **Read** and quotes the passage it relied on. The step reads **Read 1 paper · 3 passages** when full text is available, or **Read 1 paper · abstract only** when the index holds only the paper's abstract, which is common for biomedical papers.

## Go deeper on one paper

Follow up in the same chat. Bearly can read a specific paper for a specific claim, or map the work around it.

```text copy
Give me the exact effect size from the first paper.
```

```text copy
Find later papers that cite or critique that paper.
```

Ask for follow-ups, critiques, or the prior work a paper builds on. To read a whole paper rather than passages, ask Bearly to read it; for open-access papers it can bring the full text into the chat as a document.

## Turn Scholar on or off

Scholar is on by default. To change it for the current chat:

1. Select the sliders control beside the chat input.
2. Under **Search**, select **Scholar**.
3. Choose **Auto** or **Off**.

**Auto** lets Bearly use papers when they fit the question. **Off** keeps papers out of the chat entirely. To use Scholar for one message while it is off, type `/scholar`, choose it from the list, add a space, and write your question.

> **Note — When Scholar is missing**
> If **Scholar** is not listed under **Search**, a team policy may have turned it off, or Web Search may be off for your team. A team administrator can review the assigned policy under **Teams** → team name → **Policies**.


## Limits

- Coverage comes from the paper index: arXiv, bioRxiv, medRxiv, and PubMed. Journal articles outside those sources can appear through the academic web search, listed by the journal's website, without passages.
- Full-text passages are available for many arXiv papers. Papers indexed only through PubMed give Bearly the abstract, so exact figures may need the paper opened.
- Bearly does not report citation counts or rank papers by prestige. It weighs study design, sample size, replication, and recency in the answer, and says when the literature is thin or inconclusive.
- **No papers found** means the index had nothing for that phrasing. Rephrase or broaden the question, or ask for the web instead.

Paper searches run through Bearly rather than connecting your device directly to the paper index. Opening a paper is a normal visit to that website.

## Continue through the documentation

- [Documentation index](https://bearly.ai/docs.md): Browse every Bearly guide.
- [Previous: Search the web, find images, and run Deep Research](https://bearly.ai/docs/features/web-search-deep-research.md)
- [Next: Look up a person with People Search](https://bearly.ai/docs/features/people-search.md)