New CRID method enhances generative retrieval, boosting e-commerce GMV

Researchers have developed a new method called Cluster-Ranked Identifier (CRID) to improve generative retrieval systems. CRID decouples document identifiers into semantic clustering and business-value ranking, which helps avoid collisions and aligns the identifier's objective with business goals. Experiments on a large e-commerce dataset demonstrated that CRID outperforms existing embedding-based retrieval methods and resulted in a 1.06% increase in Gross Merchandise Volume (GMV) during full-traffic deployment. AI

IMPACT This new method could improve e-commerce search relevance and drive higher sales.

RANK_REASON Academic paper introducing a new method for generative retrieval. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.IR (Information Retrieval) →

• alphaXiv
• CatalyzeX
• Connected Papers
• DagsHub
• Generative Retrieval
• Gotit.pub
• Hugging Face
• Litmaps
• ScienceCast
• scite Smart Citations
• Taobao

• paper
• product
• infra

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