LLM4Rec-Learning-002: 大模型推荐的Pipline

本节是大模型LLM4Rec的Pipline,从论文、数据到源码一步步进行探索 围绕“TIGER”论文展开

主要内容

论文解读

整体结构

image-20251223094056258

RQ-VAE量化

image-20251223102407170

📌 检索不再发生在向量空间 📌 而发生在“token 空间”

量化的其他选项

  • 局部敏感哈希(LSH)
  • VQ-VAE
  • RQ-VAE

数据集

Datasets. We evaluate the proposed framework on three public real-world benchmarks from the

Amazon Product Reviews dataset[10], containing user reviews and item metadata from May 1996

to July 2014. In particular, we use three categories of the Amazon Product Reviews dataset for the

sequential recommendation task: “Beauty”, “Sports and Outdoors”, and “Toys and Games”. We

discuss the dataset statistics and pre-processing in Appendix C.

思考

  • user/item 进行embedding后进行量化,是在干什么?为什么这样做?目标?

  • 各个模块和整体模块的输入输出是什么?长啥样?