Research map / Probabilistic and scientific modeling / Molecules and proteins
Persistent Autoregressive Mapping with Traffic Rules for Autonomous Driving
Shiyi Liang, Xinyuan Chang, Changjie Wu, Huiyuan Yan, Yifan Bai, Xinran Liu, Hang Zhang, Yujian Yuan, Shuang Zeng, Mu Xu, Xing Wei
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 Map-Rule Cache (ablation) | PAMR · MapDRv2 test | HMA F1: 39.93 -> 70.37 | Improved |
| use ROI prompt instead of no prompt (ablation) | PAMR · MapDRv2 test | HMA F1: 41.74 -> 70.37 | Improved |
| use ROI prompt instead of COORD (ablation) | PAMR · MapDRv2 test | HMA F1: 56.16 -> 70.37 | Improved |
| use ROI prompt instead of BBOX (ablation) | PAMR · MapDRv2 test | HMA F1: 59.12 -> 70.37 | Improved |
| use 224x224 segments instead of 122x224 (ablation) | PAMR · MapDRv2 test | HMA F1: 53.60 -> 70.37 | Improved |
| use 224x224 segments instead of 448x224 (ablation) | PAMR · MapDRv2 test | HMA F1: 59.64 -> 70.37 | Improved |
| use 224x224 segments instead of 448x224 (vector accuracy) (ablation) | PAMR · MapDRv2 test | Vec F_vec: 0.41 -> 0.46 | Improved |
| use 10 PV images instead of 5 (ablation) | PAMR · MapDRv2 test | HMA F1: 64.66 -> 70.37 | Improved |
| use 10 PV images instead of 15 (ablation) | PAMR · MapDRv2 test | HMA F1: 64.68 -> 70.37 | Improved |
| use 10% overlap instead of 5% (ablation) | PAMR · MapDRv2 test | HMA F1: 58.20 -> 70.37 | Improved |
| use 10% overlap instead of 15% (ablation) | PAMR · MapDRv2 test | HMA F1: 51.82 -> 70.37 | Improved |
| autoregressive joint lane-rule model vs RuleVLM baseline pipeline (joint association) | PAMR · MapDRv2 test | HMA F1: 64.20 -> 70.37 | Improved |
Closest papers by meaning
- Driving by the Rules: A Benchmark for Integrating Traffic Sign Regulations into Vectorized HD Map (CVPR 2025 · similarity 0.90)
- LaneSegNet: Map Learning with Lane Segment Perception for Autonomous Driving (ICLR 2024 · similarity 0.87)
- DriveCombo: Benchmarking Compositional Traffic Rule Reasoning in Autonomous Driving (CVPR 2026 · similarity 0.87)
- Unveiling the Hidden: Online Vectorized HD Map Construction with Clip-Level Token Interaction and Propagation (NeurIPS 2024 · similarity 0.87)
- The Road Ahead in Autonomous Driving: The KITScenes Multimodal Dataset (NeurIPS 2026 · similarity 0.87)
- T2SG: Traffic Topology Scene Graph for Topology Reasoning in Autonomous Driving (CVPR 2025 · similarity 0.86)
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.
