DAP: A Discrete-token Autoregressive Planner for Autonomous Driving
This work addresses performance scaling in autonomous driving planning, offering a compact and scalable paradigm, though it appears incremental by building on autoregressive models.
The paper tackles the challenge of improving autonomous driving planning by jointly forecasting BEV semantics and ego trajectories, achieving state-of-the-art open-loop performance and competitive closed-loop results on the NAVSIM benchmark with a 160M parameter model.
Gaining sustainable performance improvement with scaling data and model budget remains a pivotal yet unresolved challenge in autonomous driving. While autoregressive models exhibited promising data-scaling efficiency in planning tasks, predicting ego trajectories alone suffers sparse supervision and weakly constrains how scene evolution should shape ego motion. Therefore, we introduce DAP, a discrete-token autoregressive planner that jointly forecasts BEV semantics and ego trajectories, thereby enforcing comprehensive representation learning and allowing predicted dynamics to directly condition ego motion. In addition, we incorporate a reinforcement-learning-based fine-tuning, which preserves supervised behavior cloning priors while injecting reward-guided improvements. Despite a compact 160M parameter budget, DAP achieves state-of-the-art performance on open-loop metrics and delivers competitive closed-loop results on the NAVSIM benchmark. Overall, the fully discrete-token autoregressive formulation operating on both rasterized BEV and ego actions provides a compact yet scalable planning paradigm for autonomous driving.