Research map / Probabilistic and scientific modeling / Molecules and proteins
Scalable End-to-End Autonomous Vehicle Testing via Rare-event Simulation
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 adaptive cross-entropy importance sampling instead of naive Monte Carlo | Rare-event evaluation in the highway simulator · Multi-agent highway | rare-event samples: 2-20 times | Helped |
| Use cross-entropy importance sampling with ρ=0.01 instead of naive Monte Carlo | Non-vision ego-vehicle policy · Multi-agent highway | dangerous scenarios sampled: 3-10 times | Helped |
| Use cross-entropy importance sampling with ρ=0.01 instead of naive Monte Carlo | Non-vision ego-vehicle policy · Multi-agent highway | variance reduction: 2-16 times variance reduction | Helped |
| Use cross-entropy importance sampling instead of naive Monte Carlo | Vision-based ego-vehicle policy · Multi-agent highway | dangerous scenarios sampled: up to 18 times | Helped |
| Decrease the quantile level ρ toward 0 (ablation) | Cross-entropy importance sampler · Non-vision ego-vehicle policy, multiple test thresholds | rare-event sampling frequency: 3-10 times | Helped |
| Select the iterate with the lowest ρ-quantile rather than the last iterate or Polyak averaging (ablation) | Cross-entropy importance sampler · Rare-event evaluation | performance | Helped |
| Reduce the per-iteration sample count to 100 (ablation) | Vision-policy cross-entropy importance sampler · Multi-agent highway, vision policy | dangerous-scenario sampling behavior | No clear effect |
Closest papers by meaning
- NAVSIM: Data-Driven Non-Reactive Autonomous Vehicle Simulation and Benchmarking (NeurIPS 2024 · similarity 0.87)
- DiffScene: Diffusion-Based Safety-Critical Scenario Generation for Autonomous Vehicles (AAAI 2025 · similarity 0.86)
- Bench2Drive: Towards Multi-Ability Benchmarking of Closed-Loop End-To-End Autonomous Driving (NeurIPS 2024 · similarity 0.86)
- Hybrid-Driving: An Autonomous Driving Decision Framework Integrating Large Language Models, Knowledge Graphs and Driving Rules (AAAI 2025 · similarity 0.85)
- Monitoring of Perception Systems: Deterministic, Probabilistic, and Learning-Based Fault Detection and Identification (Abstract Reprint) (AAAI 2024 · similarity 0.85)
- Bidirectional Motion Transformer for Safety-Critical Traffic Scenario Generation (NeurIPS 2025 · similarity 0.84)
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.