Universal Item Tokenization for Transferable Generative Recommendation
Bowen Zheng · Hongyu Lu · Yu Chen · Wayne Xin Zhao · Ji-Rong Wen
摘要
Recently, generative recommendation has emerged as a promising paradigm, attracting significant research attention. The basic framework involves an item tokenizer, which represents each item as a sequence of codes serving as its identifier, and a generative recommender that predicts the next item...