RISE: Adaptive Imagination Budgeting for World Action Models
RISE introduces a system-level adaptive imagination framework for World Action Models that replaces fixed per-scene rollout budgets with sequential Roll/Stop decisions based on expected planning benefit. It also contributes CounterDrive, a counterfactual driving dataset with expert-verified trajectory, incident, and causal annotations. Experiments on NAVSIM and nuScenes report best overall planning performance while reducing unnecessary rollout, with transfer results indicating plug-in generality.
RISE introduces a system-level controller for World Action Models that sequentially decides whether to continue imagined rollout, rather than spending a fixed imagination budget on every scene. The new action is the cost-aware Roll/Stop decision, driven by expected planning benefit.1
Context
In existing World Action Models, future world evolution is rolled out for a statically allocated number of steps. RISE replaces this with a gate that weighs expected planning improvement against computation cost. This turns imagination budget from a fixed hyperparameter into a per-scene decision.1
Assessment
The evidence for the approach rests on reported NAVSIM and nuScenes experiments plus transfer results. The strongest claim, best overall planning performance with reduced unnecessary rollout, is compatible with the abstract but cannot be weighed without numeric results or ablations.1
Caveats
This document is an arXiv abstract lacking supplementary results. The CounterDrive resource and its expert annotations are described, not demonstrated. A decisive change of mind would require released metrics, annotation reliability, and a comparison against other adaptive-budget baselines.1
Computer Science > Computer Vision and Pattern Recognition
Title:RISE: Adaptive Imagination for World Action Models
Abstract:World Action Models (WAMs) improve planning by incorporating future world evolution into action generation, yet existing methods allocate a fixed imagination budget to every scene. We propose RISE (\textbf{R}efining \textbf{I}magination through \textbf{SE}lective Rollout), a system-level adaptive imagination framework that makes sequential \textsc{Roll}/\textsc{Stop} decisions according to the expected planning benefit of continued rollout. At each step, a Latent Evaluator estimates the risk revealed by the current prefix and how much planning could improve if imagination continues, while a Rollout Gate weighs this expected benefit against additional computation cost. Since factual driving logs expose only one realized future, we further construct \textbf{CounterDrive}, a counterfactual dataset with diverse outcomes and risk levels, to enrich future dynamics and provide localized risk supervision. Each retained sample undergoes expert verification and annotation of trajectory validity, incident onset, and causal category, providing a reusable resource for safety-critical world-modeling research. Experiments on NAVSIM and nuScenes show that RISE achieves the best overall planning performance while reducing unnecessary rollout, with additional transfer results supporting its plug-in generality across WAM architectures.
Bookmark
Bibliographic and Citation Tools
Code, Data and Media Associated with this Article
Demos
Recommenders and Search Tools
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
The source text above remains the property of its publisher and is reproduced for commentary.
