一作SIGIR 2026Oral· Information Retrieval (Full)
CARD: Non-Uniform Quantization of Visual Semantic Unit for Generative Recommendation
Yibiao Wei · Jie Zou · Pengfei Zhang · Xiao Ao · Weikang Guo · Zeyu Ma · Yang Yang
摘要
Generative recommendation frameworks typically represent items as discrete Semantic IDs (SIDs). While existing studies have sought to enhance SID construction by incorporating multimodal content, collaborative signals, or more advanced quantization techniques, learning high-quality SIDs still fac...