Planning in the Backbone: DiffAdapterVLA for Native Continuous Trajectory Generation with Driving VLMs

T1Covered by 1 source · first reported by arXiv at 24 Sept, 04:00 UTC

arXiv:2609.15322v3 Announce Type: replace-cross Abstract: Pretrained driving vision-language models (VLMs) integrate visual, route, language, and driving context into rich driving priors, yet their representation objectives remain separated from continuous driving planning. Existing methods…

science

Sources

• T1Planning in the Backbone: DiffAdapterVLA for Native Continuous Trajectory Generation with Driving VLMsarXiv24 Sept, 04:00