Semantic Scholar is a free academic search engine developed by the Allen Institute for AI that uses artificial intelligence to help researchers navigate and discover relevant scientific literature. Launched publicly in 2015, it applies machine learning, natural language processing, and computer vision to analyze scientific papers and extract meaningful insights beyond traditional keyword-based search methods.

The platform indexes over 200 million research papers across all scientific disciplines, including computer science, biology, medicine, neuroscience, business, history, economics, and more. Rather than relying solely on keyword matching, Semantic Scholar understands the semantic meaning and context of research, identifying connections between papers, extracting key findings, and highlighting influential citations that might otherwise go unnoticed.

Semantic Scholar provides AI-generated summaries called TLDRs that offer quick overviews of papers, helping researchers rapidly assess relevance without reading entire documents. The system automatically extracts figures, tables, and key entities from papers, and uses citation analysis combined with influence metrics to surface the most important research. Users can create personal libraries, set up research alerts, and access Research Feeds that use machine learning to recommend papers based on their interests.

The platform offers Semantic Reader, an augmented reading experience that provides in-line citation cards with summaries and contextual information as users read. Advanced filtering allows researchers to narrow results by author, publication venue, date, and subject area. The Semantic Scholar Academic Graph API enables developers to build applications and integrate research data into their own tools and workflows.