Semantic Search
Semantic search retrieves information using meaning rather than only exact word matches. In many AI applications, it compares embeddings of a query and candidate passages to find related content.
Someone asking “How do I stop my subscription?” may need a document titled “Cancellation policy.” Embeddings can connect those phrases even though the words differ. Exact identifiers, product names, or dates may still need lexical matching or structured filters.
Semantic search selects candidates; it does not verify their claims. A passage about cancellation for a different product can be close in meaning and still be the wrong evidence. Our practical check is whether the retrieved result answers the specific question for the right entity, time, and user permissions.
Semantic search also works without a generative model: a document finder can return ranked passages directly. RAG adds a generation step that uses retrieved information to compose an answer.
Sources
- Weaviate: Hybrid search — Documents the distinction between vector similarity and keyword matching.
Go deeper
- Vector database or knowledge graph — which should you use for AI retrieval? AI, decoded · Vector Database vs. Knowledge Graph for AI Retrieval
- What is RAG, and why do AI systems use it? AI, decoded · What Is RAG (Retrieval-Augmented Generation)
- Sentence Transformers: Semantic search docs
Build query and passage embeddings and distinguish symmetric from asymmetric retrieval.
