Reproduce “From Waste to Discovery: Recycling High-Scored Unexposed Candidates in Taobao Feed 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
Ge Fan, Kai Meng, Huiping Chu, Yuning Jiang, Han Zhu, Bo Zheng
URED introduces discovery content by saving highly scored, unshown items and recalling them later when their categories have not been recently seen. It aggregates candidates over 30 days, applies a 15-day discovery lookback, and requires no retraining. In Taobao’s A/B test, discovered categories rose 3.47% and completion views 2.62%; total views rose 0.36%, not significantly. Results depend on ranker-scored supply.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.