Reproduce “SepLRM: A Foundation Model Paradigm 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
Di Chen, Xijun Xiao, Qin Ren, Weijiang Lai, Zhijian Pan, Zhentao Pan, Pengfei Nie, Cheng Yuan and 18 more
SepLRM shares user-history encoding across recommendation models, letting each query cached behavior representations for its candidates. Its encoder stores candidate-independent states. In an offline e-commerce simulation with 30 models sharing one RFM, relative ROI reaches 7.15× at a 0.46% AUC gain and 26.81× at 0.94%; this includes 30 days of cache storage, not measured serving latency.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.