Research map / Label-efficient and robust learning / Recognition and long-tail
SafeBench: A Benchmarking Platform for Safety Evaluation of Autonomous Vehicles
Chejian Xu, Wenhao Ding, Weijie Lyu, Zuxin Liu, Shuai Wang, Yihan He, Hanjiang Hu, Ding Zhao, Bo Li
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 |
|---|---|---|---|
| test on safety-critical instead of benign scenarios | TD3 (4D) · SafeBench | overall score: 0.830 -> 0.518 | Hurt |
| test on safety-critical instead of benign scenarios | PPO (4D) · SafeBench | overall score: 0.819 -> 0.622 | Hurt |
| add scenario selection (ablation) | Carla Scenario Generator · Scenario generation | overall score: 0.574 -> 0.524 | Mixed |
| use LiDAR+camera fusion (CLOCs) instead of LiDAR only (SECOND) | 3D detector · Adversarial 3D AP easy | 3D AP (%): 60.74 -> 61.98 | Improved |
| use LiDAR+camera fusion (CLOCs) instead of LiDAR only (SECOND) | 3D detector · Benign 3D AP easy | 3D AP (%): 87.31 -> 76.90 | Hurt |
| apply pose attack to PointNet++ | segmentation · Point cloud segmentation | IoU: 0.81 -> 0.40 | Hurt |
| apply point attack to Cylinder3D | segmentation · Point cloud segmentation | IoU: 0.96 -> 0.96 | No clear effect |
| switch RL algorithm from SAC to PPO under 4D | AD agent · SafeBench safety-critical | overall score: 0.497 -> 0.622 | Improved |
Closest papers by meaning
- DiffScene: Diffusion-Based Safety-Critical Scenario Generation for Autonomous Vehicles (AAAI 2025 · similarity 0.89)
- SafeAuto: Knowledge-Enhanced Safe Autonomous Driving with Multimodal Foundation Models (ICML 2025 · similarity 0.88)
- Hybrid-Driving: An Autonomous Driving Decision Framework Integrating Large Language Models, Knowledge Graphs and Driving Rules (AAAI 2025 · similarity 0.87)
- Bench2Drive: Towards Multi-Ability Benchmarking of Closed-Loop End-To-End Autonomous Driving (NeurIPS 2024 · similarity 0.86)
- Bidirectional Motion Transformer for Safety-Critical Traffic Scenario Generation (NeurIPS 2025 · similarity 0.86)
- ChatScene: Knowledge-Enabled Safety-Critical Scenario Generation for Autonomous Vehicles (CVPR 2024 · 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.
