Yutong Xu

SARL-STG: A Spatially Aware Reinforcement Learning Framework for Refining MLLMs in Spatio-Temporal Video Grounding

Spatio-Temporal Video Grounding (STVG) requires models to localize objects both spatially and temporally. Despite recent progress, existing methods struggle with complex and fine-grained spatial semantics in language descriptions, leading to error propagation from temporal to spatial grounding stage