SOURCE-LINKED INTELLIGENCE

Efficient GPU Retrieval for Semantic Search

arXiv · AI, language, vision and robotics · article · Aug 29, 2026 · UTC

Semantic Search on LinkedIn must retrieve relevant profiles from a corpus of hundreds of millions in response to natural-language queries such as "a fintech founder in Berlin who worked in payments." The deployed relevance policy is bottleneck-oriented: every active non-negotiable facet must be satisfied, and a pre-existing LLM Graded Relevance (GR) judge operationalizes this through a fixed min/median aggregation over facet grades. Cosine similarity instead averages evidence, letting a strong match on one facet mask failure on another, capping the recall of the first-stage (L0) retriever. We

Read original source ↗ Open in workspace

recordType

paper

region

Global