Research map / Label-efficient and robust learning / Domain adaptation and generalization
Interaction-Merged Motion Planning: Effectively Leveraging Diverse Motion Datasets for Robust Planning
Giwon Lee, Wooseong Jeong, Daehee Park, Jaewoo Jeong, Kuk-Jin Yoon
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 |
|---|---|---|---|
| use interaction-level instead of model-level merging (ablation) | IMMP (GameTheoretic) · SIT | ADE: 0.3687 -> 0.3380 | Improved |
| use interaction-level instead of parameter-level merging (ablation) | IMMP (GameTheoretic) · SIT | ADE: 0.3798 -> 0.3380 | Improved |
| add all-metric checkpoint selection instead of ADE-only (ablation) | IMMP (GameTheoretic) · SIT | ADE: 0.3646 -> 0.3543 | Improved |
| add intermediate-epoch checkpoints (planner and forecaster) (ablation) | IMMP (GameTheoretic) · SIT | ADE: 0.3543 -> 0.3380 | Improved |
| add fine-tuning on the target after merging (ablation) | IMMP (GameTheoretic) · SIT | ADE: 0.3380 -> 0.3157 | Improved |
| keep modules separate instead of merging two groups (ablation) | IMMP (DIPP) · SIT | ADE: 0.5666 -> 0.5112 | Improved |
| use interval C=5 instead of C=1 (ablation) | IMMP (GameTheoretic) · SIT | ADE: 0.3234 -> 0.3157 | Improved |
| use IMMP instead of Ties Merging | GameTheoretic · SIT | ADE: 1.1876 -> 0.3380 | Improved |
| use Ensemble-WTA instead of Ensemble-AVG | GameTheoretic · SIT | ADE: 0.5415 -> 0.3695 | Improved |
| add domain adaptation pretraining over target only | GameTheoretic · SIT | ADE: 0.4343 -> 0.4388 | No clear effect |
| use IMMP (no finetune) vs target only | GameTheoretic · THOR | ADE: 0.1003 -> 0.1165 | Hurt |
| add IMMP + Finetune vs ensemble with 7x cost | GameTheoretic · Zara2 | ADE: 0.4781 -> 0.4687 | Improved |
Closest papers by meaning
- SmartRefine: A Scenario-Adaptive Refinement Framework for Efficient Motion Prediction (CVPR 2024 · similarity 0.87)
- Flow Matching-Based Autonomous Driving Planning with Advanced Interactive Behavior Modeling (NeurIPS 2025 · similarity 0.86)
- UniMotion: A Unified Motion Framework for Simulation, Prediction and Planning (NeurIPS 2025 · similarity 0.86)
- Multi-Robot Motion Planning with Diffusion Models (ICLR 2025 · similarity 0.85)
- AlignDrive: Aligned Lateral-Longitudinal Planning for End-to-End Autonomous Driving (NeurIPS 2026 · similarity 0.85)
- Diffusion-Based Planning for Autonomous Driving with Flexible Guidance (ICLR 2025 · similarity 0.85)
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.
