arXiv:2611.21851v1 Announce Type: new Abstract: Autonomous driving requires anticipating how traffic will evolve while responding quickly to new observations. Video foundation models offer the first ability, but repeated history encoding and iterative video generation make them costly to use for driving. We present AR-WAM, an efficient streaming world-action model that combines autoregressive temporal modeling with single-pass future conditioning. Because its video backbone is causal, AR-WAM updates its visual context incrementally, reusing past computation and conditioning future prediction on observed motion. We further hypothesize that future features computed from recent observations can improve generalization in dynamic traffic without iterative video generation. AR-WAM therefore obtains these features in one forward pass and uses them to condition trajectory generation, and reinforcement learning then improves the policy through direct optimization of driving rewards. With a single front camera, AR-WAM achieves 91.5 PDMS after supervised training and 94.0 after reinforcement learning on NAVSIM-v1, outperforming the compared strong end-to-end world-action models. In a dynamic setting of continuous driving, supervised AR-WAM keeps its visual history across decisions and outperforms strong world-action baselines.
AR-WAM: efficient streaming world-action modeling for autonomous driving
AR-WAM, a streaming world-action model for autonomous driving detailed in arXiv paper 2611.21851v1, reached 91.5 PDMS after supervised training and 94.0 after reinforcement learning on NAVSIM-v1 using a single front camera, outperforming compared strong end-to-end world-action models. The model pairs a causal video backbone that updates visual context incrementally and reuses past computation with single-pass future conditioning, avoiding the repeated history encoding and iterative video generation that make video foundation models costly for driving. In continuous driving, supervised AR-WAM retains its visual history across decisions and beats strong world-action baselines.
AR-WAM's causal video backbone reuses past computation and conditions trajectory generation on future features from a single forward pass, scoring 91.5 PDMS supervised and 94.0 after reinforcement learning on NAVSIM-v1 with one front camera.
Run your AI side-project on zahid.host
EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.