Reproduce “Dual-disentangle Framework for Diversified Sequential 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
Haoran Zhang, Jingtong Liu, Jiangzhou Deng, Junpeng Guo
DDSRec separates a user’s history into trend-like and discrete-interest sequences, then splits each representation into category-related and category-independent signals and cross-fuses them for next-item prediction. On KuaiRec, Recall@10 is 16.1962% versus TEDDY’s 15.2948%; CE@10 rises from 1.7128 to 1.8379, and CC@10 from 0.2135 to 0.2283. These offline gains are not universal: MIND Recall@20 is 21.8538%, below TEDDY’s 22.0976%.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.