Researchers have developed STAR, a framework for predicting post-click conversion rates (PCVR) in recommender systems. STAR addresses challenges with heterogeneous features, user sequences, and target-item awareness. The framework integrates structured feature tokenization with target-aware interest representation, building on a HyFormer-style backbone. Key components include explicit user-item interaction tokens, target-aware sequence decoding, and a contrastive auxiliary objective inspired by InfoNCE. Experiments on the KDD Cup 2026 Tencent UniRec Challenge dataset demonstrated significant improvements in ranking AUC, with temporal context proving to be the most impactful element. AI
IMPACT This research could lead to more effective recommender systems, improving user experience and conversion rates in e-commerce and content platforms.
RANK_REASON The cluster contains an academic paper detailing a new framework for a specific machine learning task. [lever_c_demoted from research: ic=1 ai=1.0]
Read on arXiv cs.IR (Information Retrieval) →
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
