NAVSIM

NAVSIM is a benchmark dataset used to evaluate the closed-loop planning performance of Driving Vision-Language-Action models.

Papers using NAVSIM (5)

Drive-JEPA: Video JEPA Meets Multimodal Trajectory Distillation for End-to-End Driving2026DriveMA: Driving Vision-Language-Action Models with verifiable Meta-Actions2026NoRD: A Data-Efficient Vision-Language-Action Model that Drives without Reasoning2026FLARE: Learning Future-Aware Latent Representations from Vision-Language Models for Autonomous Driving2026SGDrive: Scene-to-Goal Hierarchical World Cognition for Autonomous Driving2026