Skip to main content

Support Explicit ANN vs. Exact Search Configuration

CCC.Vector.CN07 · Data Processing

Provide clients with the option to enforce exact-match (non-ANN) search where search fidelity is critical.

Related Capabilities

IDTitleDescription
CCC.Vector.CP02Vector IndexingThe service may be configured to build indexes optimized for similarity search, such as HNSW, IVF, or product-quantized structures, with user-selected algorithm and tuning parameters.
CCC.Vector.CP12Approximate or Exact Search ModesThe service can run a query either with an approximate nearest neighbor (ANN) algorithm for speed or with an exhaustive exact search for precision-critical applications.

Related Threats

IDTitleDescription
CCC.Vector.TH06Search Results are Manipulated via Approximate Search ErrorApproximate nearest neighbor algorithms trade recall for speed and return results that vary with index structure and tuning parameters. Vectors may be positioned so that they are omitted from, or injected into, the approximate result set while an exact search would rank them differently, allowing records to evade similarity-based detection or to bias generated responses. Integrity of search results is compromised.

Assessment Requirements

IDTextApplicability
CCC.Vector.CN07.AR01When a search request is issued, the service MUST allow the client to require exact rather than approximate results.tlp-amber, tlp-red