Policy-Grounded Dynamic Facet Suggestions for Job Search
Researchers at LinkedIn have developed a new system called dynamic facet suggestion (DFS) to improve job search relevance. DFS uses a policy-grounded, retrieval-augmented framework with a distilled small language model to offer personalized semantic attributes in real-time. This interactive mechanism helps disambiguate user intent from short, underspecified queries, leading to better job retrieval and engagement. AI
IMPACT Enhances job search relevance by providing personalized semantic attributes, potentially improving user engagement and outcomes.