GIRL-DETR: Gradient-Isolated Reinforcement Learning for Video Moment Retrieval 事件

GIRL-DETR: Gradient-Isolated Reinforcement Learning for Video Moment Retrieval arXiv:2606.00775v1 Announce Type: new Abstract: Video Moment Retrieval (VMR) task requires accurately localizing temporal boundaries aligned with natural language queries, but many models suffer from a misalignment between continuous surrogate losses and non-differentiable metrics, leading to optimization stagnation during the late stages of training and trapping boundary predictions in suboptimal solutions. Although