KP-Agent: Keyword Pruning in Sponsored Search Advertising via LLM-Powered Contextual Bandits

TL;DR

This paper introduces KP-Agent, an LLM-powered system that models keyword pruning in sponsored search advertising as a contextual bandit problem, significantly improving campaign profitability through reinforcement learning.

Contribution

It presents a novel LLM-based agentic framework with domain tools and memory for keyword pruning, addressing a previously under-explored aspect of SSA optimization.

Findings

KP-Agent achieves up to 49.28% profit increase.

Model effectively refines keyword sets via reinforcement learning.

Demonstrates practical benefits on real-world SSA data.

Abstract

Sponsored search advertising (SSA) requires advertisers to constantly adjust keyword strategies. While bid adjustment and keyword generation are well-studied, keyword pruning-refining keyword sets to enhance campaign performance-remains under-explored. This paper addresses critical inefficiencies in current practices as evidenced by a dataset containing 0.5 million SSA records from a pharmaceutical advertiser on search engine Meituan, China's largest delivery platform. We propose KP-Agent, an LLM agentic system with domain tool set and a memory module. By modeling keyword pruning within a contextual bandit framework, KP-Agent generates code snippets to refine keyword sets through reinforcement learning. Experiments show KP-Agent improves cumulative profit by up to 49.28% over baselines.