Core Concepts
Die vorgeschlagene Methode des kontrastiven Pre-Trainings verbessert die Erfassung komplexer Beziehungen in E-Commerce-Sitzungsdaten.
Stats
Session data has been widely used for understanding user behavior in e-commerce.
Extensive experiments show that UBM better captures complex intra-item semantic relations.
UBM demonstrates strong robustness when data is sparse.