
BAHE, a novel hierarchical encoding approach, decouples the representation extraction of atomic behaviors from the learning of behavior interactions, significantly improving the efficiency and effectiveness of LLM-based CTR prediction with long user sequences.


coremsg

enhancing-efficiency-and-effectiveness-of-llm-based-click-through-rate-prediction-with-long-textual-user-behaviors


Enhancing Efficiency and Effectiveness of LLM-Based Click-Through Rate Prediction with Long Textual User Behaviors
