Research map / Continual learning / Online and replay-based
NOMAD: Lifelong Trajectory Planning via Non-Parametric Bayesian Memory-Adaptive Diffusion Experts
Yixian Chen, Rufan Bai, Jiangbin Zheng, Yimin Wang, Tiantian CHEN, Wei Wang, Yuhuan Lu
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 non-parametric memory (ablation) | NOMAD · interPlan | closed-loop score: 64 -> 70 | Improved |
| add hierarchical scene encoder (ablation) | NOMAD · interPlan | closed-loop score: 68 -> 70 | Improved |
| use memory-conditioned mixture-of-experts diffusion instead of normal diffusion (ablation) | NOMAD · interPlan | closed-loop score: 67 -> 70 | Improved |
| add memory on jaywalker scenarios (ablation) | NOMAD · interPlan jaywalker | closed-loop score: 49 -> 58 | Improved |
| add memory on high-density lane changes (ablation) | NOMAD · interPlan HTD | closed-loop score: 69 -> 69 | No clear effect |
| add NOMAD memory planner vs Diffusion Planner | diffusion planner · interPlan | closed-loop score: 24 -> 70 | Improved |
| use NOMAD instead of PDM-Closed on regular driving | planner · Test14-Random (non-reactive) | CLS-NR: 90 -> 88 | Hurt |
| use NOMAD instead of SAH-Drive on Test14-Hard non-reactive | planner · Test14-Hard (non-reactive) | CLS-NR: 78 -> 78 | No clear effect |
| single shared diffusion backbone with cluster conditioning instead of planner switching | NOMAD · interPlan | inference time per frame (s): 1.89 -> 1.46 | Improved |
Closest papers by meaning
- ChainFlow-VLA: Causal Flow Planning with Vision-Language Models (NeurIPS 2026 · similarity 0.88)
- Diffusion-ES: Gradient-free Planning with Diffusion for Autonomous and Instruction-guided Driving (CVPR 2024 · similarity 0.88)
- Diffusion-Based Planning for Autonomous Driving with Flexible Guidance (ICLR 2025 · similarity 0.87)
- Future-Aware End-to-End Driving: Bidirectional Modeling of Trajectory Planning and Scene Evolution (NeurIPS 2025 · similarity 0.87)
- DriveFuture: Future-Aware Latent World Models for Autonomous Driving (NeurIPS 2026 · similarity 0.87)
- Learning to Drive in New Cities Without Human Demonstrations (NeurIPS 2026 · 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.
