Research map / Label-efficient and robust learning / Recognition and long-tail
Decoupling Scene Perception and Ego Status: A Multi-Context Fusion Approach for Enhanced Generalization in End-to-End Autonomous Driving
Jiacheng Tang, Mingyue Feng, Jiachao Liu, Yaonong Wang, Jian Pu
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 dual-branch decoupling and fusion (without auxiliary regularizers) (ablation) | AdaptiveAD baseline · nuScenes val | avg L2 (m): 0.57 -> 0.62 | Hurt |
| add dual-branch decoupling and fusion (without auxiliary regularizers) (ablation) | AdaptiveAD baseline · nuScenes val | avg collision rate (%): 0.22 -> 0.15 | Improved |
| add BEV unidirectional distillation (ablation) | AdaptiveAD · nuScenes val | avg collision rate (%): 0.15 -> 0.08 | Improved |
| add BEV unidirectional distillation (ablation) | AdaptiveAD · nuScenes val | avg L2 (m): 0.62 -> 0.58 | Improved |
| add scene-aware initialization of fusion query (ablation) | AdaptiveAD · nuScenes val | avg L2 (m): 0.58 -> 0.52 | Improved |
| add autoregressive online mapping (ablation) | AdaptiveAD · nuScenes val | avg L2 (m): 0.52 -> 0.47 | Improved |
| add autoregressive online mapping (ablation) | AdaptiveAD · nuScenes val | avg collision rate (%): 0.12 -> 0.12 | No clear effect |
| replace deformable attention with path attention (ablation) | AdaptiveAD · nuScenes val | avg collision rate (%): 0.15 -> 0.12 | Improved |
| add path attention plug-in | UniAD · nuScenes val | avg L2 (m): 0.73 -> 0.68 | Improved |
| add autoregressive online mapping plug-in | SparseDrive · nuScenes val | avg L2 (m): 0.58 -> 0.53 | Improved |
| add autoregressive online mapping plug-in | SparseDrive · nuScenes val | avg collision rate (%): 0.06 -> 0.06 | No clear effect |
| use AdaptiveAD instead of VAD under zeroed ego velocity | AdaptiveAD vs VAD · nuScenes val | avg L2 (m): 5.54 -> 4.08 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- PPAD — PPAD: Iterative Interactions of Prediction and Planning for End-to-end Autonomous Driving (ECCV 2024) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- DualAD: Disentangling the Dynamic and Static World for End-to-End Driving (CVPR 2024 · similarity 0.88)
- Prioritizing Perception-Guided Self-Supervision: A New Paradigm for Causal Modeling in End-to-End Autonomous Driving (NeurIPS 2025 · similarity 0.88)
- Transfer Your Perspective: Controllable 3D Generation from Any Viewpoint in a Driving Scene (CVPR 2025 · similarity 0.87)
- SGDrive: Scene-to-Goal Hierarchical World Cognition for Autonomous Driving (CVPR 2026 · similarity 0.87)
- Intend, Reflect, Refine: An Adaptive Multimodal Reflection Framework for Autonomous Driving (NeurIPS 2026 · similarity 0.86)
- DriveTransformer: Unified Transformer for Scalable End-to-End Autonomous Driving (ICLR 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.
