Reproduce “Think Globally, Generate Personally: Multi-View Thinking with Adaptive Generation for Generative 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
Yuxuan Hu, Yingyi Zhang, Wenlin Zhang, Chao Zhang, Haotian Nan, Ziwei Liu, Jiancheng Dong, Tongzhou Wu and 6 more
MV-Thinker recommends next items by generating discrete semantic-ID tokens in sequence. It combines full history and recent behavior, then selects a nested residual-adapter path to transform decoder states through a shared vocabulary head. On Instruments with LETTER, H@5 is 0.0906 versus 0.0882 for LETTER+GEMs; MV-Thinker leads listed metrics across four datasets and two backbones, with baselines varying by setting.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.