RISE
2026 Datasets
Dataset Analysis
RISE adaptively decides when to stop imagination rollout in World Action Models, using a Latent Evaluator and Rollout Gate, with CounterDrive counterfactual data; outperforms on NAVSIM and nuScenes.
Provenance
Collected from papers.
Derived from paper: RISE: Adaptive Imagination for World Action Models
Related papers
RISE: Adaptive Imagination for World Action ModelsActive Inference as the Test-Time Scaling Law for Physical AI AgentsBoltzmann-GPT: Bridging Energy-Based World Models and Language GenerationDiscrete-WAM: Unified Discrete Vision-Action Token Editing for World-Policy LearningDreamSAC: Learning Hamiltonian World Models via Symmetry ExplorationDrivingGen: A Comprehensive Benchmark for Generative Video World Models in Autonomous DrivingGeoStream: Toward Precise Camera Controlled Streaming Video GenerationNous: A Predictive World Model for Long-Term Agent MemoryOmni-WorldBench: Towards a Comprehensive Interaction-Centric Evaluation for World ModelsOne Lens, Many Worlds : A Capability-Typed Interface for World-Model InterpretabilityPhysicsMind: Sim and Real Mechanics Benchmarking for Physical Reasoning and Prediction in Foundational VLMs and World ModelsRiding the Shifting Potential: When Reactive Control Suffices for Multi-Goal BehaviorSimulating clinical interventions with a generative multimodal model of human physiologyToward World Modeling of Physiological Signals with Chaos-Theoretic Balancing and Latent DynamicsWhat Makes Video World Model Latents Action-Relevant: Prediction over ReconstructionWorld Action Models: The Next Frontier in Embodied AIWorld Model for Robot Learning: A Comprehensive SurveyYoCausal: How Far is Video Generation from World Model? A Causality PerspectiveDeep SPI: Safe Policy Improvement via World ModelsEmbodied World Models Emerge from Navigational Task in Open-Ended EnvironmentsGEM: A Generalizable Ego-Vision Multimodal World Model for Fine-Grained Ego-Motion, Object Dynamics, and Scene Composition ControlGLAM: Global-Local Variation Awareness in Mamba-based World ModelLearning an Adversarial World Model for Automated Curriculum Generation in MARLOccProphet: Pushing Efficiency Frontier of Camera-Only 4D Occupancy Forecasting with Observer-Forecaster-Refiner FrameworkThe brain-AI convergence: Predictive and generative world models for general-purpose computationToward Memory-Aided World Models: Benchmarking via Spatial ConsistencyWorld-in-World: World Models in a Closed-Loop WorldWorld4Drive: End-to-End Autonomous Driving via Intention-aware Physical Latent World ModelYume-1.5: A Text-Controlled Interactive World Generation ModelCarDreamer: Open-Source Learning Platform for World Model based Autonomous DrivingCausal World Representation in the GPT ModelGenie: Generative Interactive EnvironmentsGrounded Answers for Multi-agent Decision-making Problem through Generative World ModelLearning World Models for Unconstrained Goal NavigationOccLLaMA: An Occupancy-Language-Action Generative World Model for Autonomous DrivingScaling Laws for Pre-training Agents and World ModelsPhysiFormer: Learning to Simulate Mechanics in World SpaceWorldOdysseyBench: An Open-World Benchmark for Long-Horizon Stability of Interactive World ModelsWorldRoamBench: An Open-World Benchmark for Long-Horizon Stability of Interactive World ModelsGeographic Diversity Beats Data Volume for Cross-Domain Generalization in Zero-Label JEPA Driving World ModelsThinking in Video: Can Video Generators Really Reason About the Real World?Planning with Transformers: Chain of Computation and Structured Context WindowsApple-π: Benchmarking Thinking with Video Towards Law-Grounded Physical IntelligenceTemporal-Distance JEPA: Plan-Aware Representation Learning for Latent World Model Predictive ControlVisualPatchWorld: Code World Models as Latent Structured Representations for PlanningStateFlow: Building, Evolving, and Accessing 3D World States for PrevisualizationGenerating Long Videos of Dynamic ScenesVideoGPA: Distilling Geometry Priors for 3D-Consistent Video Generation