Researchers have developed LCM-Promo, a specialized Large Commerce Model (LCM) finetuned from a large language model using transactional commerce data. This model predicts promotion response by learning monotonic curves and conversion-threshold probabilities from production logs. In a real-world A/B test for churn reactivation, LCM-Promo demonstrated a 1% increase in conversion rates and a 6% reduction in promotion costs compared to the existing Segment-EMA baseline. AI
IMPACT This LLM application demonstrates a practical method for optimizing marketing spend and improving conversion rates in e-commerce.
RANK_REASON The item describes a research paper detailing a new LLM-based model for promotion allocation. [lever_c_demoted from research: ic=1 ai=0.7]
Read on Hugging Face Daily Papers →
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →