InVDriver: Intra-Instance Aware Vectorized Query-Based Autonomous Driving Transformer

End - to - end autonomous driving with its holistic optimization capabilities, has gained increasing tr action in academia and industry . Vectorized representations, which preserve instance - level topological information while reducing computational overhead, have emerged as a promising paradigm. While existing vectorized query - based frameworks often overlook the inherent spatial correlatio ns among intra - instance points, resulting in geometrically inconsistent outputs (e.g., fragmented HD map elements or oscillatory trajectories). To address these limitations, we propose InVDriver, a novel vectorized query - based system that systematically mo dels intra - instance spatial dependencies through masked self - attention layers, thereby enhancing planning acc uracy and trajectory smoothness . Across all core modules -- perception, prediction, and planning -- InVDriver incorporates masked self - attention mechanis ms that restrict attention to intra - instance point interactions, enabling coordinated refinement of structural elements while suppressing irrelevant inter - instance noise. Experimental results on the nuScenes benchmark demonstrate t hat InVDriver achieves sta te - of - the - art performance, surpassing prior methods in both accuracy and safety, while maintainin g high computational efficiency . Our work validates that explicit modeling of intra - instance geometric coherence is critical for advancing vectorized autonomou s driving systems, bridging the gap between theoretical advantages of end - to - end frameworks and practical deployment requirements.