Mode-as-Sequence: Translating Multimodal Motion Prediction into Unified Sequential Mode Modeling 事件

Mode-as-Sequence: Translating Multimodal Motion Prediction into Unified Sequential Mode Modeling arXiv:2605.24037v1 Announce Type: new Abstract: Multimodal motion forecasting is inherently under-supervised: each training scene provides only one realized future, yet multiple plausible futures exist. This sparse supervision often leads to mode collapse (redundant hypotheses and insufficient mode coverage) and unreliable confidence ranking when predicting a small set of trajectories. We propose Mo