Xiaomi Auto World Model 2.0: Causal World Models for Real-World Driving and Embodied Intelligence
cs.AI, cs.LG
Submitted: 2026-11-01
Updated: 2026-11-01
Code: https://github.com/NVIDIA/flashdreams
Project page: https://xiaomiautoworldmodel.github.io/CausalWorldModel
Terminology
Sources
- Cosmos World Foundation Model Platform for Physical AI
- Cosmos 3: Omnimodal World Models for Physical AI
- NVIDIA OmniDreams: Real-Time Generative World Model for Closed-Loop Autonomous Vehicle Simulation
- AgiBot World Colosseo: A Large-scale Manipulation Platform for Scalable and Intelligent Embodied Systems
- NuPlan: A closed-loop ML-based planning benchmark for autonomous vehicles
- SANA-Video 2.0: Hybrid Linear Attention with Attention Residuals for Efficient Video Generation
- FPD-Drive: Future Policy Distillation for End-to-End Autonomous Driving
- RoboTwin 2.0: A Scalable Data Generator and Benchmark with Strong Domain Randomization for Robust Bimanual Robotic Manipulation
- DGGT: Feedforward 4D Reconstruction of Dynamic Driving Scenes using Unposed Images
- PointForward: Feedforward Driving Reconstruction through Point-Aligned Representations
- Open X-Embodiment: Robotic Learning Datasets and RT-X Models
- LIBERO-Plus: In-depth Robustness Analysis of Vision-Language-Action Models
- MagicDrive-V2: High-Resolution Long Video Generation for Autonomous Driving with Adaptive Control
- DreamDojo: A Generalist Robot World Model from Large-Scale Human Videos
- CDCP: Conditional Diffusion Model with Contextual Prompts for Multi-task Offline Safe Reinforcement Learning
- Genesis: Multimodal Driving Scene Generation with Spatio-Temporal and Cross-Modal Consistency
- Learning to Retrieve: Internalizing Memory Retrieval for Video World Models
- CoGen: 3D Consistent Video Generation via Adaptive Conditioning for Autonomous Driving
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents