Tag
This paper examines how the underrepresentation of elderly riders in mobility datasets introduces systematic bias into mobility modeling, using Citi Bike data from Jersey City. It shows that models trained on majority-dominated populations misrepresent elderly mobility behavior, and that higher-capability models do not necessarily improve subgroup fidelity under limited demographic data.
WebWatcher is a multimodal agent for deep research that uses synthetic trajectories and reinforcement learning to achieve superior performance in complex visual and textual information retrieval tasks. The paper also introduces BrowseComp-VL, a new benchmark for evaluating multimodal agents.