Beyond Successful Robot Demonstrations: Learning from Human Interactions and Modeling Failure Dynamics
cs.RO, cs.CV
Submitted: 2026-10-09
Updated: 2026-10-09
Project page: https://peiyantu.github.io/HF-WM
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- Wan: Open and Advanced Large-Scale Video Generative Models
- Cosmos 3: Omnimodal World Models for Physical AI
- World Action Models are Zero-shot Policies
- Fast-WAM: Do World Action Models Need Test-time Future Imagination?
- $\omega$-EVA: Envision, Verify, and Act with Latent Interactive World Models
- PiL-World: A Chunk-Wise World Model for VLA Policy-in-the-Loop Evaluation
- SC3-Eval: Evaluating Robot Foundation Models via Self-Consistent Video Generation
- Robotic Video World Models: A Survey of Applications, Research Challenges, Future Directions
- DreamDojo: A Generalist Robot World Model from Large-Scale Human Videos
- MiraBench: Evaluating Action-Conditioned Reliability in Robotic World Models
- PointWorld: Scaling 3D World Models for In-The-Wild Robotic Manipulation
- ABot-PhysWorld: Interactive World Foundation Model for Robotic Manipulation with Physics Alignment
- EgoSim: Egocentric World Simulator for Embodied Interaction Generation
- GTA: Advancing Image-to-3D World Generation via Geometry Then Appearance Video Diffusion
- Embody4D: A Generalist Data Engine for Embodied 4D World Modeling
- TesserAct: Learning 4D Embodied World Models
- TrajectoryCrafter: Redirecting Camera Trajectory for Monocular Videos via Diffusion Models
- NeoVerse: Enhancing 4D World Model with in-the-wild Monocular Videos
- Ctrl-World: A Controllable Generative World Model for Robot Manipulation
- CoWTracker: Tracking by Warping instead of Correlation
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