A Cascaded Generative Approach for e-Commerce Recommendations

TL;DR

This paper presents a cascaded generative framework for e-commerce storefronts that enhances personalization, semantic cohesion, and adaptability, resulting in a 2.7% increase in cart adds per page view.

Contribution

It introduces a novel two-stage generative approach for storefront construction, combining theme and keyword generation with fine-tuning and evaluation frameworks.

Findings

Achieved +2.7% lift in cart adds per page view.

Demonstrated that fine-tuned models approach closed-weight LLM performance.

Developed scalable AI-driven content evaluation and filtering methods.

Abstract

Personalized storefronts in large e-commerce marketplaces are often assembled from many independent components: static themes per page section ("placement"), retrieval systems to fetch eligible products per placement, and pointwise rankers to order content. While effective in optimizing for aggregate preferences, this paradigm is rigid and can limit personalization and semantic cohesion across the page. This makes it poorly suited to support dynamic objectives and merchandising requirements over time. To address this, we introduce a cascaded merchandising framework that decomposes storefront construction into two generative tasks: (i) placement-level theme generation and (ii) constrained keyword generation per placement to power product retrieval. Teacher-student fine-tuning is leveraged to improve scalability of this framework under production latency and cost constraints. Fine-tuned model ablations are shown to approach closed-weight LLM performance. We further contribute frameworks for AI-driven content evaluation and quality filtering, enabling safe and automated deployment of dynamic content at scale. Generative output is fused with traditional ranking models to preserve hybrid infrastructure. In online experiments, this framework yields an estimated +2.7% lift in cart adds per page view over a strong baseline.