Chao-Hua Yang

Bridge the Unseen Gap: Enhancing Non-overlapping Cross-domain CTR Prediction via Profile Retrieval

This paper proposes LLM-PRIT, a framework for Large Language Model-generated Profile Retrieval & Instance Transfer, which utilizes an LLM as a universal semantic interpreter to generate domain-agnostic, transferable profiles for users and items, encapsulating open-world knowledge.