Reproduce “Revisit Recommender System in the Permutation Prospective” 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
Yufei Feng, Yu Gong, Fei Sun, Junfeng Ge, Wenwu Ou
The paper presents PRS, a re-ranking system that searches a bounded set of promising product orders and scores candidate lists with a model conditioned on their displayed order. In Taobao’s Minidetail recommendation scenario, PRS improved items browsed by 11.0% and clicked by 8.7% over production, with 7.3 milliseconds average inference cost; these are reported aggregate online lifts, not proof of optimality.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.