Research map / Probabilistic and scientific modeling / Molecules and proteins
Perceiving the Near, Reasoning the Distant: Coherent Long-Horizon Trajectory Prediction for Autonomous Driving
Hua Hu, Zikang Zhou, Qian Zhou, Zihao Wen, Junjie Hu, Xinhong Chen, Zhengmin Jiang, Yung-Hui Li, Jianping 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 |
|---|---|---|---|
| decouple decoding into near and distant stages (vs one-shot) (ablation) | NDPNet baseline · Argoverse 2 val | minFDE6: 1.264 -> 1.203 | Improved |
| add Near-to-Distant Temporal Bridging (ablation) | dual-stage NDPNet · Argoverse 2 val | minFDE6: 1.203 -> 1.173 | Improved |
| add MAC loss (ablation) | dual-stage NDPNet + bridging · Argoverse 2 val | minAHE6: 0.079 -> 0.047 | Improved |
| add MAC loss (ablation) | SceneTransformer · Argoverse 2 | minFDE6: 2.49 -> 2.39 | Improved |
| add MAC loss (ablation) | HiVT · Argoverse 2 | minFDE6: 1.98 -> 1.94 | Improved |
| add MAC loss (ablation) | QCNet · Argoverse 2 | minAHE6: 0.10 -> 0.04 | Improved |
| use three stages instead of two (ablation) | NDPNet · Argoverse 2 val | minFDE6: 1.165 -> 1.193 | Hurt |
| dual-stage decoding without bridging or MAC (ablation) | one-shot baseline · Argoverse 2 val | minAHE6: 0.083 -> 0.085 | No clear effect |
| near-future window of 2s vs 1s or 3s (ablation) | NDPNet · Argoverse 2 val | minFDE6: 1.259 (1s) / 1.253 (2s) / 1.260 (3s) | Improved |
| near window 4s vs 2s (ablation) | NDPNet · WOMD val | minFDE6: 1.101 -> 1.134 | Hurt |
| dual-stage design vs one-shot (ablation) | NDPNet · Argoverse 2 val | FPS: 15.0 -> 13.1 | Hurt |
Closest papers by meaning
- OneVL: One-Step Latent Reasoning and Planning with Vision-Language Explanation (NeurIPS 2026 · similarity 0.89)
- DriveFuture: Future-Aware Latent World Models for Autonomous Driving (NeurIPS 2026 · similarity 0.88)
- Future-Aware End-to-End Driving: Bidirectional Modeling of Trajectory Planning and Scene Evolution (NeurIPS 2025 · similarity 0.88)
- ChainFlow-VLA: Causal Flow Planning with Vision-Language Models (NeurIPS 2026 · similarity 0.87)
- LaKD: Length-agnostic Knowledge Distillation for Trajectory Prediction with Any Length Observations (NeurIPS 2024 · similarity 0.87)
- Don't Shake the Wheel: Momentum-Aware Planning in End-to-End Autonomous Driving (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.
