Skip to main content
2026-07-07
v0.1.7: Entities in Collections

New: Entities in the knowledge graph

Collections now extract named entities (people, organizations, products) from ingested content, in addition to claims and predictions. Entities become nodes in the collection’s knowledge graph, connected to every claim or prediction that mentions them — turning previously disconnected facts about the same person, company, or product into a single connected cluster.

Entity types

TypeDescription
PERSONA specific named individual
ORGANIZATIONA specific named company, institution, or group
PRODUCTA specific named brand or model (generic categories are not extracted)

Changed endpoints

EndpointChange
GET /api/v1/collections/{id}/graphResponse nodes may now have "type": "ENTITY" in addition to CLAIM/PREDICTION
GET /api/v1/collections/{id}/contextAdds matched_entities (entity names/aliases found in the question) and entity_facts (claims/predictions linked to those entities)
POST /api/v1/collections/{id}/qa-verifyGrounding now combines vector search with entity-graph traversal. If a question and a claim share a named entity but are worded very differently, the claim is now found — improving verification accuracy in that case

Why this matters

Previously, retrieval relied only on vector similarity between the question and ingested text. If a question mentioned an entity by name but phrased things differently from the source content, relevant claims could be missed. Entities give the graph a second, exact-match retrieval path alongside semantic search.