Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Divide-and-Conquer for Lane-Aware Diverse Trajectory Prediction
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use DAC instead of EWTA for multimodal hypothesis training | CPI trajectory prediction · CPI test set | Avg FDE: 5.76 -> 5.58 | Helped |
| Use DAC instead of WTA for multimodal hypothesis training | CPI trajectory prediction · CPI test set | Avg EMD: 1.79 -> 1.31 | Helped |
| Use lane-anchor ALAN predictions ranked by BofA | nuScenes trajectory prediction · nuScenes validation | mADE_10: 1.24 -> 1.10 | Helped |
| Use lane-anchor ALAN predictions ranked by BofA | nuScenes trajectory prediction · nuScenes validation | OffRoadRate: 0.07 -> 0.01 | Helped |
| Use ALAN (BofA) instead of MHA_JAM | nuScenes trajectory prediction · nuScenes validation | Miss_2_10: 45 -> 45 | No clear effect |
| Add DAC to multi-head DeepLabV3 | DeepLabV3 semantic segmentation · Pascal2012 val set | mIoU: 82.14 -> 83.44 | Helped |
| Use hypercolumn descriptors instead of global context (ablation) | ALAN · nuScenes validation ablation | mADE_10: 1.88 -> 1.33 | Helped |
| Add reference centerline input and predict in xy coordinates (ablation) | ALAN · nuScenes validation ablation | mADE_10: 1.88 -> 1.27 | Helped |
| Use lane anchors and predict in nt coordinates without auxiliary xy predictions (ablation) | ALAN · nuScenes validation ablation | mADE_10: 1.27 -> 1.45 | Hurt |
| Add auxiliary xy trajectory predictions to lane-anchor nt prediction (ablation) | ALAN · nuScenes validation ablation | mADE_10: 1.45 -> 1.41 | Helped |
| Add cross-coordinate regularization to anchor and auxiliary predictions (ablation) | ALAN · nuScenes validation ablation | mADE_10: 1.41 -> 1.17 | Helped |
| Use DAC instead of EWTA with regularized lane-anchor predictions (ablation) | ALAN · nuScenes validation ablation | mADE_10: 1.20 -> 1.17 | Helped |
Closest papers by meaning
- AlignDrive: Aligned Lateral-Longitudinal Planning for End-to-End Autonomous Driving (NeurIPS 2026 · similarity 0.87)
- LaneSegNet: Map Learning with Lane Segment Perception for Autonomous Driving (ICLR 2024 · similarity 0.87)
- Density-Adaptive Model Based on Motif Matrix for Multi-Agent Trajectory Prediction (CVPR 2024 · similarity 0.87)
- GoIRL: Graph-Oriented Inverse Reinforcement Learning for Multimodal Trajectory Prediction (ICML 2025 · similarity 0.87)
- LaKD: Length-agnostic Knowledge Distillation for Trajectory Prediction with Any Length Observations (NeurIPS 2024 · similarity 0.87)
- UniMotion: A Unified Motion Framework for Simulation, Prediction and Planning (NeurIPS 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.