Reproduce “MotiveMob: Motivation as Semantic Action for Closed-Loop Human Mobility Generation” 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
Mengkun Gao, Zengqing Wu, Renhe Jiang, Jiawei Wang, Yusong Wang, Chuang Yang, Shuyuan Zheng, Makoto Onizuka and 1 more
MotiveMob generates daily human mobility trajectories by predicting a motivation proxy before each next place and arrival time. User context and recent history inform predictions, and proposed visits undergo time and speed checks. On Tokyo data, it reduces unseen-user time-by-activity divergence by 18.3% on average versus the best baselines; this does not validate actual intent or individual routes.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.