Humata
ProductivitySearch, summarize and compare uploaded documents with source-linked answers. Humata adds team-oriented document access and page-based plans for ongoing knowledge work.
Questions across filesCited passages
RESEARCH DISCOVERY / FIND THE PAPER BEHIND THE CLAIM
Semantic Scholar
A free AI-assisted scholarly search service from Ai2, combining paper discovery, citation connections, personal research organization and reading assistance on supported papers.
Official Semantic Reader example showing AI-assisted skimming highlights; feature availability varies by paperExplore the image ↓
What you can do with it
What this tool is for
Semantic Scholar is an AI-driven academic search and discovery project within Ai2. It organizes scholarly records and uses machine learning to surface relationships and useful context. It is particularly helpful when you know a topic or a seed paper but need a manageable route into the surrounding literature.
Its role is discovery and reading assistance. A useful result may still be a preliminary study, an outdated method or a paper that answers a different question. Treat summaries and citation signals as ways to choose what to read, then evaluate the original source before incorporating its claims into your work.
01
Begin a literature review
02
Check a technical claim
03
Follow an active topic
04
Prepare a reading group
A closer look at the product
Official product image. Click to inspect the details.
Product interface and examples
Key features, practical tasks
Capabilities in practice
- Scholarly search: Look for a paper, author or research topic and refine the candidate set. A narrow query with the core method or problem often produces a more reviewable starting list than a broad subject label.
- Citation connections: Move between a paper, its references and later citing work. Use those connections to trace a research discussion, while remembering that a citation does not necessarily endorse the cited result.
- TLDR summaries where available: The official FAQ describes short AI-generated paper summaries with limited domain coverage. Use them for initial triage, then read the abstract and relevant sections for scope and qualifications.
- Library organization: An account can save papers for later reading. Build a small folder structure around questions or projects so a growing collection remains useful rather than becoming an unreviewed archive.
- Research feeds and alerts: Account features include recommendations and notifications. Tune them using relevant examples and review the resulting stream instead of assuming all suggested papers meet your inclusion criteria.
- Semantic Reader: Supported papers can display contextual citation information and reading aids. Feature availability varies; do not assume every PDF in the index has the same enhanced reading experience.
- Developer resources: The Academic Graph API and datasets support research applications under their own access and usage conditions. Website access does not automatically imply unlimited programmatic throughput.
Get started, step by step
- 01
Write a specific search question
- 02
Screen paper records
- 03
Open the original evidence
- 04
Save a structured reading note
A practical first workflow
A practical first workflow
- Write a specific search question. Include the phenomenon, method or application you care about. Keep a note of alternative terminology used by the field so one wording does not limit discovery.
- Screen paper records. Check title, authors, date and abstract. Separate papers worth reading from those that are merely adjacent to the topic.
- Open the original evidence. Follow available full-text or publisher links and inspect methods, results and limitations. Use Semantic Reader when offered, while checking generated reading aids against the source.
- Save a structured reading note. Record the question answered, evidence used and remaining uncertainties. Preserve the source link and bibliographic details for later verification.
Example prompt or task: Find papers addressing this specific research question. For each promising result, record its method, evaluation setting and limitation after checking the original paper; keep uncertain claims separate from verified findings.
This is an editorial research workflow, not a claim that the search service performs a complete systematic review or validates every scientific result.
Find the right plan for you
Free research service
Explore official plans ↗Free Publisher full-text access can be separate
Search and discovery are free. Finding a paper record does not remove a publisher paywall or guarantee an available PDF.
What is free and what can cost extra
Semantic Scholar describes its search and discovery tools as free. Personal organization and notification features use an account, while the underlying scholarly records can be explored without treating account creation as a paid subscription.
Full-text availability depends on publishers, open-access copies and institutional access. API keys, rate limits and dataset terms are separate from ordinary browsing. This listing does not promise free access to every paper or unrestricted reuse of every indexed document.
A few things worth knowing
Frequently asked questions
Is Semantic Scholar free?
Its scholarly search and discovery service is free. Publisher access can be separate.
Do I need an account to search?
Basic paper access does not require an account; saving, feeds and alerts use account features.
Are all PDFs free?
No. Some records lead to publisher or institutional access options, and some lack a full-text link.
Does every paper have a TLDR?
No. Availability and domain coverage are limited.
Is Semantic Reader available everywhere?
No. The enhanced reader and individual features apply to supported papers.
Does a high citation count prove quality?
No. Read the paper and the context of the citations.
Can developers use the data?
Official APIs and datasets exist, with their own conditions and limits.
The details behind your decision
Explore platforms, inputs and outputs, licensing, and access requirements.
Four useful starting projects
Four useful starting projects
- Begin a literature review: Search a concrete question and identify a small seed set. Record why each paper is relevant before expanding through references and citations.
- Check a technical claim: Find the paper behind a quoted result and inspect the evaluation setting. Compare the original conclusion with the way it is described in a blog or presentation.
- Follow an active topic: Save useful papers and configure suitable account features yourself. Review new recommendations periodically and remove material that drifts away from the research question.
- Prepare a reading group: Choose a paper with clear relevance, inspect related work and draft questions about assumptions and limitations. Share source links so participants can examine the same evidence.
Platforms, inputs, outputs and setup
Platforms, inputs, outputs and setup
The service is accessed through a browser. Search results and paper records can lead to publisher pages, PDFs or metadata-only records, so the amount of evidence available differs by paper.
Semantic Reader is documented for supported papers, especially parts of the arXiv collection, with some features limited further by language or field. An account library can provide context for personalized features, but it does not make the coverage universal.
Data, permissions and licensing
Data, permissions and licensing
Search topics and saved libraries can reveal research interests. Review account, alert and sharing settings according to the sensitivity of your work, and distinguish a private reading workflow from publicly shared material.
AI-generated summaries and reading aids can contain errors. Verify important claims in the original text, retain attribution and respect the separate rights attached to full-text documents.
Sources, documentation and help
Official sources and review date
Official sources and review date
Reviewed October 3, 2026. Product facts come from the official sources below. Suggested projects, prompts and review methods are FindGoodAI editorial guidance, not measured performance results.
Your experience can help someone choose
COMMUNITY
Sign in to rate and comment
Tried it? Share your experience
Loading experiences…
Manage my comments
Community and moderation
Saved tools are private. Approved comments are public; edits return to moderation. Comment counts include approved comments and replies. Each person has one rating. New ratings are automatically approved; withdrawn, unapproved or invalid ratings do not count. Share real experiences and avoid spam, private information or personal attacks.
For moderation appeals or data requests: support@findgoodai.com
NEXT STEP
Explore Semantic Scholar ↗Start with a focused question and inspect the evidence
Bring a real task and see how it fits the way you work.
Jump to a section
User experiences