OMS: On-the-fly, Multi-Objective, Self-Reflective Ad Keyword Generation via LLM Agent
This addresses keyword decision challenges for advertisers in sponsored search advertising, representing an incremental improvement over existing LLM-based methods.
The paper tackles the problem of automated keyword generation for sponsored search advertising by addressing limitations of existing LLM-based methods, proposing OMS which requires no training data and optimizes keywords based on multiple performance metrics, with experiments showing it outperforms existing methods on benchmarks and real-world campaigns.
Keyword decision in Sponsored Search Advertising is critical to the success of ad campaigns. While LLM-based methods offer automated keyword generation, they face three major limitations: reliance on large-scale query-keyword pair data, lack of online multi-objective performance monitoring and optimization, and weak quality control in keyword selection. These issues hinder the agentic use of LLMs in fully automating keyword decisions by monitoring and reasoning over key performance indicators such as impressions, clicks, conversions, and CTA effectiveness. To overcome these challenges, we propose OMS, a keyword generation framework that is On-the-fly (requires no training data, monitors online performance, and adapts accordingly), Multi-objective (employs agentic reasoning to optimize keywords based on multiple performance metrics), and Self-reflective (agentically evaluates keyword quality). Experiments on benchmarks and real-world ad campaigns show that OMS outperforms existing methods; ablation and human evaluations confirm the effectiveness of each component and the quality of generated keywords.