Yaqing Wang

Revealing memory utilization patterns through attention can transform how agents refine their memory, leading to substantial gains in performance and efficiency.

Today's best smartphone GUI agents stumble when faced with the messy reality of personalized user workflows, achieving only limited success on a new benchmark designed to mimic real-world use.

By explicitly encoding Green's function theory into a neural architecture, DGNet learns spatiotemporal PDEs from just tens of training trajectories, outperforming existing methods.

LLMs learn faster and perform better in decision-making tasks when rewarded for being uncertain, not just for succeeding.