ConDenseDrive: Knowledge-Dense, Token-Sparse Visual Encoder for End-to-End Autonomous Driving

This paper presents ConDenseDrive, a knowledge-dense and token-sparse visual encoder for end-to-end autonomous driving. The three-stage framework distills visual, semantic, and geometric knowledge from foundation models, integrates spatiotemp, and applies token compression to improve inference speed and planning performance. It achieves 94.5 PDMS on NAVSIM v1 and 90.8 EPDMS on NAVSIM v2, with a 37.4 HD-Score in zero-shot evaluation on HUGSIM.