REVIEW 14 cited by
DrivingWorld: Constructing World Model for Autonomous Driving via Video GPT
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Recent successes in autoregressive (AR) generation models, such as the GPT series in natural language processing, have motivated efforts to replicate this success in visual tasks. Some works attempt to extend this approach to autonomous driving by building video-based world models capable of generating realistic future video sequences and predicting ego states. However, prior works tend to produce unsatisfactory results, as the classic GPT framework is designed to handle 1D contextual information, such as text, and lacks the inherent ability to model the spatial and temporal dynamics essential for video generation. In this paper, we present DrivingWorld, a GPT-style world model for autonomous driving, featuring several spatial-temporal fusion mechanisms. This design enables effective modeling of both spatial and temporal dynamics, facilitating high-fidelity, long-duration video generation. Specifically, we propose a next-state prediction strategy to model temporal coherence between consecutive frames and apply a next-token prediction strategy to capture spatial information within each frame. To further enhance generalization ability, we propose a novel masking strategy and reweighting strategy for token prediction to mitigate long-term drifting issues and enable precise control. Our work demonstrates the ability to produce high-fidelity and consistent video clips of over 40 seconds in duration, which is over 2 times longer than state-of-the-art driving world models. Experiments show that, in contrast to prior works, our method achieves superior visual quality and significantly more accurate controllable future video generation. Our code is available at https://github.com/YvanYin/DrivingWorld.
Forward citations
Cited by 14 Pith papers
-
Adaptive-WAM: Quality-Guided Early-Exit Planning from Intermediate Video-Diffusion Features
An early-exit, quality-guided planner on a Wan2.2-5B video-diffusion backbone reaches 90.8 PDMS on NAVSIM in about 170 ms, without generating future video at deployment.
-
Orbis 2: A Hierarchical World Model for Driving
A hierarchical driving world model — planning in compressed DINO space at 2 Hz and rendering detailed frames at 10 Hz — achieves state-of-the-art long-horizon stability, steering response, and representation quality.
-
Instant NuRec: Feed-Forward 3D Gaussian Reconstruction for Driving Scene Simulation
A feed-forward model reconstructs a layered, simulation-ready 3D Gaussian world from multi-view driving video in ~1.5 s, with quality approaching per-scene optimized reconstruction.
-
OpenLongTail: Generative Scaling of Long-Tail Driving Data
Pose-informed diffusion with Plücker rays, depth warps, and cross-view memory converts monocular long-tail videos into multi-view assets that improve closed-loop driving robustness nearly to ground-truth multi-view levels.
-
Thinking Ahead: Foresight Intelligence in MLLMs and World Model
FSU-QA is a new nuScenes-based VQA benchmark for foresight reasoning; current VLMs score about 41-49% accuracy, and fine-tuning an 8B model on it outperforms GPT-5 and Gemini on the benchmark.
-
OmniNWM: Omniscient Driving Navigation World Models
OmniNWM jointly generates long panoramic multi-modal driving videos, controls them precisely via normalized Plücker ray-maps, and derives dense driving rewards from generated 3D occupancy.
-
A Comprehensive Survey on World Models for Embodied AI
A unified three-axis taxonomy — functionality, temporal modeling, spatial representation — organizes the world-model literature for embodied AI.
-
Epona: Autoregressive Diffusion World Model for Autonomous Driving
An autoregressive diffusion world model jointly generates the next camera frame and a multi-step trajectory, enabling long videos and real-time planning for autonomous driving.
-
GeoDrive: 3D Geometry-Informed Driving World Model with Precise Action Control
GeoDrive conditions a frozen video diffusion model on a 3D-rendered version of the requested ego trajectory, cutting trajectory-following error by 42% versus Vista while using 99.7% less training data.
-
UNIVERSE: Unified Video Action Models for Autonomous Driving with Flexible Mask-Modulated Modality Generation
A single mask-modulated DiT that co-trains future video and trajectories yields stronger autonomous-driving action generalization and 4.3× faster trajectory-only inference than dual-DiT designs.
-
3D and 4D World Modeling: A Survey
A survey that defines 3D/4D world modeling, organizes methods into VideoGen, OccGen, and LiDARGen categories, and compiles datasets, metrics, and benchmark numbers.
-
Depth Anything at Any Condition
A fine-tuned Depth Anything V2 model using perturbation consistency and spatial distance constraints improves monocular depth estimation under adverse conditions without any labeled data.
-
Step-Audio-AQAA: a Fully End-to-End Expressive Large Audio Language Model
Step-Audio-AQAA, a 130B end-to-end audio language model using dual-codebook tokens, text-audio interleaving, masked DPO and weight merging, is claimed to outperform Kimi-Audio and Qwen-Omni on the authors' StepEval-Au...
-
MGVQ: Could VQ-VAE Beat VAE? A Generalizable Tokenizer with Multi-group Quantization
Splitting quantization across multiple small sub-codebooks with nested masking raises VQ-VAE reconstruction fidelity, giving MGVQ rFID 0.49 and PSNR 24.70 on ImageNet at 16 times downsampling.
Discussion (0). Continue with ORCID to comment.