Reproduce “TriGen: A Generative Paradigm for Trigger-Induced 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
Muyu Zou, Haibo Xing, Shunye Tang, Liyan Tie, Lingyu Mu, Yandong Wang, Jinxin Hu
TriGen guides follow-up recommendation generation by injecting a clicked item into recent history, adaptively combining trigger-focused and history-preserving representations, and conditioning each step of semantic-ID decoding. On an industrial dataset, it reports Recall@10 of 15.50 versus 12.15 for trigger append. In a six-market product-detail-page trial, revenue rose 3.27% and GMV 0.90%; results are specific to this platform and setting.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.