← Back to the feed
AI & softwareResearch findings

LEADRE: LLM-Powered Multi-Faceted Knowledge-Based Advertisement Recommender System

Original published 2024.11.21 11:22 KSTAdded to Blackdeck 2026.10.25 08:50 KST

What has changed?

Intent-aware prompt engineering integrates multi-faceted knowledge to tailor ads to user interests

Key points

  1. LEADRE leverages LLM's multi-faceted knowledge to reflect user interests in ad recommendations
  2. Introduces intent-aware prompting and DPO to align LLM outputs with ad semantics and business value
  3. Efficiently deploys in online environments with latency-tolerant and latency-sensitive services

Connect it to your work

LLM generates personalized ads by aligning with user intent and ad semantics

What remains unproven

This research note is based on an abstract; check the original for test conditions and application scope.

Originals used for this note

Links open the original publisher in a new tab.

  1. arxivLEADRE: Multi-Faceted Knowledge Enhanced LLM Empowered Display Advertisement Recommender System