Research paper

RTSM: Knowledge Distillation with Diverse Signals for Efficient Real-Time Semantic Matching in E-Commerce

Research from AmazonPublished by Association for Computational LinguisticsApr 2025PDF

Explore how a smaller model can learn from several signals to match products and queries fast enough for real-time search.

RTSM: Knowledge Distillation with Diverse Signals for Efficient Real-Time Semantic Matching in E-Commerce cover

Resource access

Get the full resource

Enter your business details to read the full resource.

First nameLast nameWork emailCompanyJob titleCountry or region

Sanjay Agrawal, Vivek Sembium. RTSM: Knowledge Distillation with Diverse Signals for Efficient Real-Time Semantic Matching in E-Commerce. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 3: Industry Track). © 2025 Association for Computational Linguistics. CC BY 4.0. 10.18653/v1/2025.naacl-industry.2. Original PDF, unmodified.

CC BY 4.0

RTSM: Knowledge Distillation with Diverse Signals for Efficient Real-Time Semantic Matching in E-Commerce

Related research.

Leveraging Product Catalog Patterns for Multilingual E-commerce Product Attribute Prediction

ShopperBench: A Benchmark for Personalized Shopping with Persona-Guided Simulation

When to Defer: Scope-Aware Cascades for Structured Attribute Generation in E-commerce Product Catalogs