Reproduce “SIDRec: End-to-End Semantic ID Representation and Modeling for Recommendation” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Zhuohang Jiang, Yongsen Pan, Pangjing Wu, Zihuai Zhao, Zheng Hu, Yuxin Chen
SIDRec builds item codes for recommendation by combining fixed multimodal features with collaborative embeddings learned from a ranking model. It updates the codes during click-prediction training, using a tokenizer and a model that represents user histories. On three public datasets, it achieves the highest AUC and GAUC among 15 CTR baselines; these are offline benchmark results.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.