Policy-Grounded Dynamic Facet Suggestions for Job Search

TL;DR

This paper introduces a real-time, personalized facet suggestion system for job search that improves user intent understanding and search outcomes through a policy-grounded, retrieval-augmented ranking framework.

Contribution

It presents a novel dynamic facet suggestion method combining offline taxonomy curation, embedding retrieval, and small language model scoring for improved job search relevance.

Findings

High precision in offline evaluation of suggestions

Significant online engagement improvements

Enhanced job search outcomes in A/B tests

Abstract

Job seekers often initiate search with short, underspecified queries. At LinkedIn, over 80% of job-related queries contain three or fewer keywords, making accurate user intent inference and relevant job retrieval particularly challenging. We present dynamic facet suggestion (DFS), an interactive query refinement mechanism that facilitates intent disambiguation by surfacing personalized semantic attributes conditioned on the joint user-query context in real time. We propose a policy-grounded, retrieval-augmented ranking framework for facet suggestion, comprising offline taxonomy curation, embedding-based retrieval of top-K candidates, and distilled small language model (SLM) based candidate scoring. The system is optimized for real-time serving via pointwise single-token scoring with batching and prefix caching. Offline evaluation demonstrates high precision for generated suggestions, and online A/B tests show significant improvements in suggestion engagement and job search outcomes.