LLMs boost App Store search relevance, increasing conversion rates

Researchers have developed a method to enhance search relevance by using Large Language Models (LLMs) to generate textual relevance labels. A specialized, fine-tuned LLM proved more effective than a larger pre-trained model for this task. By generating millions of these labels, they augmented the App Store's ranking system, resulting in a 0.24% increase in conversion rate and improved performance on less common queries. AI

IMPACT LLM-generated labels can significantly improve search result relevance and user conversion rates, particularly for niche queries.

RANK_REASON Academic paper detailing a novel application of LLMs to improve search relevance and a specific metric (conversion rate). [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• App Store
• Evangelia Christakopoulou
• Large Language Models

• paper
• product

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →