A search technique that matches content based on meaning rather than exact keyword overlap.
Semantic search, powered by embeddings, can find a relevant document even if it doesn't contain the exact words you searched for — because it understands conceptual similarity, not just text matching.