Reproduce “BrowseFlow: Sparse Event Flow Matching for Generative User Simulation” 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

Yitian Pang, Jiayi Zhang, Zhenjia Ma, Xiaowei Yue, Dagui Chen, Jian Xu

BrowseFlow models a recommendation-feed response as browsing extent and ordered sparse key events, such as clicks. It learns response distributions by sampling extent and generating events with conditional insertion flow, rather than every scrolling action. On TaobaoBrowse, its Energy Score (0.07264) and Count W1 (0.0697) are lower than the best baselines, supporting improved measured distributional fit, not true individual response probabilities.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.