Research map / Video / Video generation
DriveLaW: Unifying Planning and Video Generation in a Latent Driving World
Tianze Xia, Yongkang Li, Lijun Zhou, Jingfeng Yao, Kaixin Xiong, Haiyang Sun, Bing Wang, Kun Ma, Guang Chen, Hangjun Ye, Wenyu Liu, Xinggang Wang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| condition planner on video-generator latents instead of BEV features (ablation) | DriveLaW-Act · NAVSIM Navtest | PDMS: 84.1 -> 89.1 | Improved |
| condition planner on video latents instead of VLM hidden states (ablation) | DriveLaW-Act · NAVSIM Navtest | PDMS: 86.5 -> 89.1 | Improved |
| add driving-domain video pretraining (7.6M samples) (ablation) | DriveLaW video generator · NAVSIM Navtest | PDMS: 85.9 -> 89.1 | Improved |
| scale video pretraining from 76k to 3.8M samples (ablation) | DriveLaW video generator · NAVSIM Navtest | PDMS: 87.0 -> 87.8 | Improved |
| use latents from later denoising step (t=5) rather than t=1 (ablation) | DriveLaW-Act · NAVSIM Navtest | PDMS: 89.1 -> 86.9 | Hurt |
| use latents from denoising step t=10 (ablation) | DriveLaW-Act · NAVSIM Navtest | PDMS: 89.1 -> 23.2 | Hurt |
| add first-stage training on long low-res clips (ablation) | DriveLaW-Video · nuScenes validation | FVD: 109.3 -> 81.3 | Improved |
| add second-stage high-res training (ablation) | DriveLaW-Video · nuScenes validation | FVD: 93.2 -> 81.3 | Improved |
| add noise reinjection (ablation) | DriveLaW-Video · nuScenes validation | FVD: 102.1 -> 81.3 | Improved |
| add noise reinjection (ablation) | DriveLaW-Video · nuScenes validation | FID: 6.1 -> 4.6 | Improved |
| chain video generation into planner instead of parallel design | DriveLaW vs Epona · NAVSIM Navtest | PDMS: 86.2 -> 89.1 | Improved |
| longer horizon generation (100 frames) versus Epona | DriveLaW-Video · NuPlan FVD100 | FVD: 277.3 -> 296.1 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Epona — Epona: Autoregressive Diffusion World Model for Autonomous Driving (ICCV 2025) · better on 7 of 8 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Driving into the Future: Multiview Visual Forecasting and Planning with World Model for Autonomous Driving (CVPR 2024 · similarity 0.91)
- From Forecasting to Planning: Policy World Model for Collaborative State-Action Prediction (NeurIPS 2025 · similarity 0.91)
- Driving in the Occupancy World: Vision-Centric 4D Occupancy Forecasting and Planning via World Models for Autonomous Driving (AAAI 2025 · similarity 0.90)
- DriveFuture: Future-Aware Latent World Models for Autonomous Driving (NeurIPS 2026 · similarity 0.90)
- Future-Aware End-to-End Driving: Bidirectional Modeling of Trajectory Planning and Scene Evolution (NeurIPS 2025 · similarity 0.89)
- UniMotion: A Unified Motion Framework for Simulation, Prediction and Planning (NeurIPS 2025 · similarity 0.89)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
