Research map / Image generation / Fast diffusion sampling
DiffMM: Efficient Method for Accurate Noisy and Sparse Trajectory Map Matching via One Step Diffusion
Chenxu Han, Sean Bin Yang, Jilin Hu
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 Transformer encoder in point representation (vs FFN) (ablation) | DiffMM · Beijing r=0.1 | accuracy (%): 84.89 -> 85.39 | Improved |
| add Transformer encoder in point representation (vs FFN) (ablation) | DiffMM · Beijing r=0.5 | accuracy (%): 90.06 -> 90.32 | Improved |
| add candidate-segment attention (vs mean) (ablation) | DiffMM · Beijing r=0.1 | accuracy (%): 82.71 -> 85.39 | Improved |
| use shortcut model (vs traditional diffusion) in one step (ablation) | DiffMM · Beijing r=0.1 | accuracy (%): 83.53 -> 85.39 | Improved |
| use one-step shortcut diffusion instead of baselines at sparse sampling | DiffMM · Beijing r=0.1 | accuracy (%): 68.25 -> 85.39 vs DeepMM | Improved |
| DiffMM vs HMM at dense sampling | DiffMM · Porto r=0.2 | accuracy (%): 92.46 -> 93.43 | Improved |
| use one-step DiffMM inference instead of baselines | DiffMM · Beijing r=0.1 | inference time per 1000 trajectories (s): 20.57 -> 1.18 | Improved |
| DiffMM training time vs DeepMM | DiffMM · Beijing r=0.1 | training time per epoch (min): 9.07 vs 10.66 | No clear effect |
| increase training set size from 16,000 to 128,000 (ablation) | DiffMM · Porto r=0.1 | accuracy (%): 86.01 -> 90.03 | Improved |
Closest papers by meaning
- PLMTrajRec: A Scalable and Generalizable Trajectory Recovery Method with Pre-trained Language Models (NeurIPS 2025 · similarity 0.86)
- TrajMamba: An Efficient and Semantic-rich Vehicle Trajectory Pre-training Model (NeurIPS 2025 · similarity 0.86)
- Generative Human Trajectory Recovery via Embedding-Space Conditional Diffusion (ICML 2025 · similarity 0.85)
- ProDiff: Prototype-Guided Diffusion for Minimal Information Trajectory Imputation (ICML 2025 · similarity 0.85)
- Bridging Traffic State and Trajectory for Dynamic Road Network and Trajectory Representation Learning (AAAI 2025 · similarity 0.84)
- Metric Flow Matching for Smooth Interpolations on the Data Manifold (NeurIPS 2024 · similarity 0.84)
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.
