Research map / 3D vision / 3D generation
SparseWorld: A Flexible, Adaptive, and Efficient 4D Occupancy World Model Powered by Sparse and Dynamic Queries
Chenxu Dang, Haiyan Liu, Jason Bao, Pei An, Xinyue Tang, An Pan, Jie Ma, Bingchuan Sun, Yan 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 |
|---|---|---|---|
| add Adaptive Scaling of query ranges (ablation) | SparseWorld · Occ3D-nuScenes | Avg mIoU: 11.45 -> 11.82 | Improved |
| add temporal mask in TS-MHSA (ablation) | SparseWorld · Occ3D-nuScenes | Avg mIoU: 11.52 -> 11.82 | Improved |
| add 4D position encoding in SCF (ablation) | SparseWorld · Occ3D-nuScenes | Avg mIoU: 11.58 -> 11.82 | Improved |
| add ego state condition (ablation) | SparseWorld · Occ3D-nuScenes | Avg mIoU: 11.31 -> 11.82 | Improved |
| use Spatial instead of common cross-attention (no ego state) (ablation) | SparseWorld · Occ3D-nuScenes planning | L2 (m): 1.25 -> 1.01 | Improved |
| use Spatial instead of common cross-attention (no ego state) (ablation) | SparseWorld · Occ3D-nuScenes | mIoU: 10.28 -> 10.64 | Improved |
| use self-scheduled timestamps instead of none (ablation) | SparseWorld · Occ3D-nuScenes | Avg mIoU: 10.95 -> 11.82 | Improved |
| use self-scheduled timestamps instead of manual (ablation) | SparseWorld · Occ3D-nuScenes | Avg mIoU: 11.37 -> 11.82 | Improved |
| SparseWorld vs PreWorld on forecasting | SparseWorld · Occ3D-nuScenes | Avg mIoU: 9.06 -> 13.20 | Improved |
| SparseWorld vs OccLLaMA-F on IoU | SparseWorld · Occ3D-nuScenes | Avg IoU: 22.99 -> 22.03 | Hurt |
| sparse queries instead of dense grids | SparseWorld · Occ3D-nuScenes | FPS: 1.0 -> 8.0 | Improved |
Later papers that compared with this method
1 later accepted paper reports SparseWorld as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- SparseWorld-TC: Trajectory-Conditioned Sparse Occupancy World Model (CVPR 2026 · better on 1 of 1 shared result)
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
- SparseOcc: Rethinking Sparse Latent Representation for Vision-Based Semantic Occupancy Prediction (CVPR 2024 · similarity 0.89)
- Driving in the Occupancy World: Vision-Centric 4D Occupancy Forecasting and Planning via World Models for Autonomous Driving (AAAI 2025 · similarity 0.88)
- Spatiotemporal Decoupling for Efficient Vision-Based Occupancy Forecasting (CVPR 2025 · similarity 0.88)
- UnO: Unsupervised Occupancy Fields for Perception and Forecasting (CVPR 2024 · similarity 0.88)
- Semi-Supervised Vision-Centric 3D Occupancy World Model for Autonomous Driving (ICLR 2025 · similarity 0.87)
- OccMamba: Semantic Occupancy Prediction with State Space Models (CVPR 2025 · similarity 0.87)
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.
