Semantic Scholar

Semantic Scholar is a free academic search engine built by Ai2 (the Allen Institute for AI). It indexes more than 200 million papers from every field of science and uses its own machine-learning models to help researchers find, triage and read literature faster. There is no paid tier: search, the library, alerts and the API are free.

Finding papers

• Search with filters for field, journal or conference, author, publication type and date range.
• TLDRs. One-sentence AI summaries of a paper's objective and results appear in search results, for tens of millions of papers in computer science, biology and medicine.
• Highly Influential Citations. A model flags the citing papers that a work genuinely shaped, so you can start with the citations that matter instead of wading through all of them.
• Cite any paper in BibTeX, MLA, APA or Chicago.

Keeping up

• Library and Research Feeds. Save papers into folders, then switch on an AI Research Feed per folder to get recommendations that improve as you rate papers relevant or not.
• Alerts email you about new papers by an author or new citations of a paper.

Reading

Semantic Reader is an augmented PDF reader for most arXiv papers. It shows citation cards with TLDRs in-line where a reference appears, marks citations connected to your own library, adds AI skimming highlights labelled Goal, Method and Result, and explains terms in context.

For developers

The Semantic Scholar API and open datasets expose papers, authors and citations for research tools and literature-review pipelines.

Who it's for

Students and researchers running a literature review, and anyone who needs a free, citation-aware alternative to general web search for scientific work.