The reviewed record of science sign in
Pith

arxiv: 2506.09042 · v3 · pith:DZDIMZX7 · submitted 2025-06-10 · cs.CV

Cosmos-Drive-Dreams: Scalable Synthetic Driving Data Generation with World Foundation Models

Reviewed by Pith T0 review T1 audit T2 compute T3 formal T4 kernel pith:DZDIMZX7record.jsonopen to challenge →

classification cs.CV
keywords drivingdatachallengingcosmoscosmos-drive-dreamsgenerationmodelsnvidia
0
0 comments X
read the original abstract

Collecting and annotating real-world data for safety-critical physical AI systems, such as Autonomous Vehicle (AV), is time-consuming and costly. It is especially challenging to capture rare edge cases, which play a critical role in training and testing of an AV system. To address this challenge, we introduce the Cosmos-Drive-Dreams - a synthetic data generation (SDG) pipeline that aims to generate challenging scenarios to facilitate downstream tasks such as perception and driving policy training. Powering this pipeline is Cosmos-Drive, a suite of models specialized from NVIDIA Cosmos world foundation model for the driving domain and are capable of controllable, high-fidelity, multi-view, and spatiotemporally consistent driving video generation. We showcase the utility of these models by applying Cosmos-Drive-Dreams to scale the quantity and diversity of driving datasets with high-fidelity and challenging scenarios. Experimentally, we demonstrate that our generated data helps in mitigating long-tail distribution problems and enhances generalization in downstream tasks such as 3D lane detection, 3D object detection and driving policy learning. We open source our pipeline toolkit, dataset and model weights through the NVIDIA's Cosmos platform. Project page: https://research.nvidia.com/labs/toronto-ai/cosmos_drive_dreams

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 27 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. LongLive-RAG: A General Retrieval-Augmented Framework for Long Video Generation

    cs.CV 2026-06 unverdicted novelty 7.0

    LongLive-RAG formulates long video generation as retrieval-augmented generation by treating self-generated latents as a dynamic searchable history and adding a Window Temporal Delta Loss for better retrieval.

  2. Closed Loop Dynamic Driving Data Mixture for Real-Synthetic Co-Training

    cs.CV 2026-05 unverdicted novelty 7.0

    AutoScale is a closed-loop data engine using Graph-RAE for scene representation and Cluster-GA for importance-based retrieval to improve real-synthetic co-training for autonomous driving.

  3. HorizonDrive: Self-Corrective Autoregressive World Model for Long-horizon Driving Simulation

    cs.CV 2026-05 conditional novelty 7.0

    HorizonDrive enables stable long-horizon autoregressive driving simulation via anti-drifting teacher training with scheduled rollout recovery and teacher rollout distillation.

  4. ScenarioControl: Vision-Language Controllable Vectorized Latent Scenario Generation

    cs.CV 2026-04 unverdicted novelty 7.0

    ScenarioControl introduces the first vision-language controllable generator for realistic vectorized 3D driving scenarios with temporal consistency across actor views.

  5. Off the Rails: Hijacking the Scoring Head in Generative End-to-End Driving Planners with Safety-Violating Adversarial Perturbations

    cs.RO 2026-06 unverdicted novelty 6.0

    Derail adversarial perturbations hijack the scoring head in generative E2E driving planners, flipping safe to unsafe trajectory selection with 39-80% score drops and up to 50% collision rates.

  6. DriveJudge: Rethinking Autonomous Driving Evaluation with Vision-Language Models

    cs.CV 2026-06 unverdicted novelty 6.0

    DriveJudge combines VLM reasoning with rule functions on a new 33,577-sample human-annotated dataset, outperforming EPDMS by 21.23 AUC on quality classification and DriveCritic by 6.5% on trajectory preference.

  7. Dash2Sim: Closed-Loop Driving Simulation from in-the-wild Dashcam Videos

    cs.CV 2026-06 unverdicted novelty 6.0

    Dash2Sim recovers metric geo-referenced 4D scenes from in-the-wild monocular dashcam videos to enable the ROADWork4D benchmark, revealing that current closed-loop planners fail on work zone lane changes.

  8. Geometry-Aware Implicit Memory for Video World Models

    cs.CV 2026-06 unverdicted novelty 6.0

    GIM-World adds a camera-queryable geometry distillation head and pruning rule to implicit memory in video world models, claiming better long-horizon geometric consistency on the MIND benchmark than explicit and implic...

  9. Gamma-World: Generative Multi-Agent World Modeling Beyond Two Players

    cs.CV 2026-05 unverdicted novelty 6.0

    A multi-agent video world model using simplex rotary agent encoding and sparse hub attention achieves better fidelity, controllability, and consistency than baselines while generalizing from 2 to 4 players.

  10. E$^3$C: Video Generation with 3D Environmental Memory and Ego-Exo Human Pose Control

    cs.CV 2026-05 unverdicted novelty 6.0

    E³C is a video diffusion model that disentangles persistent 3D scene structure via point-cloud memory from human dynamics via ego-exo pose controls for improved egocentric video generation on the Nymeria dataset.

  11. AnyScene: Towards Highly Controllable Driving Scene Generation at Anywhere and Beyond

    cs.RO 2026-05 unverdicted novelty 6.0

    AnyScene is an occupancy-centric framework using a Spatial-Temporal Occupancy Diffusion Transformer and Geometry-Grounded View Expansion to generate controllable driving scenes and videos from BEV layouts.

  12. Sensor2Sensor: Cross-Embodiment Sensor Conversion for Autonomous Driving

    cs.CV 2026-05 unverdicted novelty 6.0

    Sensor2Sensor converts in-the-wild monocular dashcam videos into high-fidelity multi-modal AV sensor data using 4D Gaussian Splatting to synthesize training pairs and a diffusion model for the cross-embodiment translation.

  13. Sensor2Sensor: Cross-Embodiment Sensor Conversion for Autonomous Driving

    cs.CV 2026-05 unverdicted novelty 6.0

    Sensor2Sensor uses 4D Gaussian Splatting to create synthetic training pairs and a diffusion model to convert monocular dashcam videos into high-fidelity multi-modal AV sensor data.

  14. HorizonDrive: Self-Corrective Autoregressive World Model for Long-horizon Driving Simulation

    cs.CV 2026-05 unverdicted novelty 6.0

    HorizonDrive is a new anti-drifting autoregressive training and distillation method that enables minute-scale stable driving video rollouts by making the teacher model rollout-capable via scheduled rollout recovery an...

  15. Human Cognition in Machines: A Unified Perspective of World Models

    cs.RO 2026-04 unverdicted novelty 6.0

    The paper introduces a unified framework for world models that fully incorporates all cognitive functions from Cognitive Architecture Theory, highlights under-researched areas in motivation and meta-cognition, and pro...

  16. Video Generation Models as World Models: Efficient Paradigms, Architectures and Algorithms

    eess.IV 2026-03 unverdicted novelty 6.0

    Video generation models can function as world simulators if efficiency gaps in spatiotemporal modeling are bridged via organized paradigms, architectures, and algorithms.

  17. DriveLaW:Unifying Planning and Video Generation in a Latent Driving World

    cs.CV 2025-12 unverdicted novelty 6.0

    DriveLaW unifies video world modeling and trajectory planning by injecting video-generator latents into a diffusion planner, achieving SOTA video prediction and a new record on the NAVSIM planning benchmark.

  18. Ctrl-World: A Controllable Generative World Model for Robot Manipulation

    cs.RO 2025-10 unverdicted novelty 6.0

    A controllable world model trained on the DROID dataset generates consistent multi-view robot trajectories for over 20 seconds and improves generalist policy success rates by 44.7% via imagined trajectory fine-tuning.

  19. MoWorld: A Flash World Model

    cs.CV 2026-07 conditional novelty 5.0

    MoWorld achieves up to 50 FPS real-time interactive world simulation on NPUs by combining a 3D-native data engine, curriculum cross-frame pretraining, autoregressive distillation, and mixed-precision parallel inference.

  20. Creating Impactful Autonomous Driving Datasets: A Strategic Guide from Research Gap to Benchmark

    cs.CV 2026-07 unverdicted novelty 5.0

    A framework for creating impactful autonomous driving datasets is presented, starting with gap diagnosis between data and evaluation problems and selecting minimal operators, demonstrated with the KITScenes dataset.

  21. FrozenDrive: Zero-Shot Text-Guided Driving Scene Generation and Data Augmentation with Parameter-Free Frozen Diffusion Model

    cs.CV 2026-06 unverdicted novelty 5.0

    FrozenDrive enables zero-shot text-guided generation of consistent multi-view driving scenes via a parameter-free frozen diffusion backbone with spatio-temporal attention, improving autonomous driving models on advers...

  22. Agentic Environment Engineering for Large Language Models: A Survey of Environment Modeling, Synthesis, Evaluation, and Application

    cs.CL 2026-06 unverdicted novelty 5.0

    This survey categorizes agentic environments for LLMs by eight attributes and domains, introduces symbolic and neural synthesis paradigms with evaluation, and outlines four agent evolution pathways plus three environm...

  23. ViPE: Video Pose Engine for 3D Geometric Perception

    cs.CV 2025-08 unverdicted novelty 5.0

    ViPE estimates camera intrinsics, motion, and dense near-metric depth from uncalibrated videos, outperforming baselines on TUM and KITTI while releasing annotations for 96M frames across real and generated videos.

  24. NVIDIA OmniDreams: Real-Time Generative World Model for Closed-Loop Autonomous Vehicle Simulation

    cs.CV 2026-06 unverdicted novelty 4.0

    OmniDreams is a real-time generative world model mid- and post-trained from the Cosmos diffusion model on 21k hours of driving data to autoregressively generate action-conditioned videos for closed-loop AV simulation.

  25. Advancing Reliable Synthetic Video Detection: Insights from the SAFE Challenge

    cs.CV 2026-05 unverdicted novelty 4.0

    The SAFE challenge shows measurable progress in detecting synthetic videos across different generators but persistent weaknesses against post-processing operations.

  26. Advancing Open-source World Models

    cs.CV 2026-01 unverdicted novelty 4.0

    LingBot-World is presented as an open-source world model that delivers high-fidelity simulation, minute-level contextual consistency, and real-time interactivity under one second latency.

  27. World Simulation with Video Foundation Models for Physical AI

    cs.CV 2025-10 unverdicted novelty 4.0

    Cosmos-Predict2.5 unifies text-to-world, image-to-world, and video-to-world generation in one model trained on 200M clips with RL post-training, delivering improved quality and control for physical AI.