FutureDuet: Decoupling Observation Access from Future Supervision in World Action Models

arXiv:2609.34362 · cs.RO · Submitted 2026-09-28 · Read on arXiv

cs.RO

Submitted: 2026-09-28

Updated: 2026-09-28

Comments: 13 pages, 7 figures. Project page: https://1723578110.github.io/futureduet-web/

Project page: https://1723578110.github.io/futureduet-web/ABSTRACT

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Benchmark — 10× in this paper · explained in 2 episodes
• World Model — 6× in this paper · explained in 7 episodes
• Decoupling — 5× in this paper · explained in 2 episodes
• Alignment — 4× in this paper · explained in 1 episode
• Foundation Model — 4× in this paper · explained in 3 episodes
• Hanging Mug — 4× in this paper
• Turn Switch — 4× in this paper
• Open Microwave — 3× in this paper
• Adjust Bottle — 2× in this paper
• Blocks Ranking RGB — 2× in this paper
• Fine-Tuning — 2× in this paper · explained in 5 episodes
• Generalization — 2× in this paper · explained in 13 episodes
• Grab Roller — 2× in this paper
• Transformer — 2× in this paper · explained in 2 episodes
• Ablation — 1× in this paper · explained in 1 episode
• Benchmarking — 1× in this paper · explained in 3 episodes
• Diffusion Model — 1× in this paper · explained in 8 episodes
• Embeddings — 1× in this paper · explained in 2 episodes
• Flow Matching — 1× in this paper · explained in 8 episodes
• Foundation Models — 1× in this paper · explained in 7 episodes
• Mask Generation — 1× in this paper · explained in 2 episodes
• Self-Supervised Learning — 1× in this paper · explained in 7 episodes

Sources

• Motus: A Unified Latent Action World Model
• RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control
• SAM 3: Segment Anything with Concepts
• RynnVLA-002: A Unified Vision-Language-Action and World Model
• WorldVLA: Towards Autoregressive Action World Model
• GR-2: A Generative Video-Language-Action Model with Web-Scale Knowledge for Robot Manipulation
• LaWAM: Latent World Action Models for Efficient Dynamics-Aware Robot Policies
• Vidar: Embodied Video Diffusion Model for Generalist Manipulation
• Unified 4D World Action Modeling from Video Priors with Asynchronous Denoising
• Point Policy: Unifying Observations and Actions with Key Points for Robot Manipulation
• Foresight Without Seeing: Latent Futures for World Action Models
• DreamGen: Unlocking Generalization in Robot Learning through Video World Models
• Cosmos Policy: Fine-Tuning Video Models for Visuomotor Control and Planning
• Causal World Modeling for Robot Control
• Video Generators are Robot Policies
• Mixture-of-Transformers: A Sparse and Scalable Architecture for Multi-Modal Foundation Models
• JEPA-WAM: Learning Vision-Language-Action Policies with Joint-Embedding World Modeling
• DiT4DiT: Jointly Modeling Video Dynamics and Actions for Generalizable Robot Control
• V-JEPA 2.1: Unlocking Dense Features in Video Self-Supervised Learning
• GR00T N1: An Open Foundation Model for Generalist Humanoid Robots

Related papers

• FMT x: An Efficient and Asymptotically Optimal Extension of the Fast Marching Tree for Dynamic Replanning
• MPCFormer: A physics-informed data-driven approach for explainable socially-aware autonomous driving
• RoboLab: A High-Fidelity Simulation Benchmark for Analysis of Task Generalist Policies
• ARC: A Reasoning Recipe for Robot Foundation Models
• HRDexDB: A 4D Dexterous Grasping Dataset Across Human and Multiple Robot Embodiments
• APT: Action Expert Pretraining Improves Instruction Generalization of Vision-Language-Action Policies