Embedding
An embedding represents content as a numeric vector. The model and comparison method determine which similarities become visible.
An embedding is a numeric representation of content in a vector space. For text, a model creates these representations so that, for example, similarities between a question and document passages can be compared.
A search example
In a hypothetical knowledge portal, someone asks about “sending back” an item while the instructions use “returns”. Suitable embeddings can help find the related passage despite different wording. Whether that works depends on the model and data.
High similarity proves neither identical meaning in every detail nor factual correctness. It grants no reading permission either. Selection, validity and permitted access to a result need separate consideration.
Sources
Related terms
- Vector databaseA vector database stores and searches vector representations. A similar match is not automatically a valid or permitted answer source.
- Retrieval-augmented generationRAG combines information retrieval with generated answers. Access to sources does not remove the need for quality and permission design.
- Confidence scoreA confidence score assesses a model prediction according to a particular method. It does not grant business approval.
- Model cardA model card documents an AI model’s purpose, foundations, evaluation conditions and known limitations for a particular version.
- AI readinessAI readiness concerns the foundations of an AI project: objectives, usable data, skills, responsibilities and an operating workflow.
- Context engineeringContext engineering prepares information for individual model calls: relevant instructions, messages, sources and current states.
Articles on this topic
AI & Automation with ASES
AI assistants and automated workflows