Research map / Multi-agent systems and games / Human-AI and goal-conditioned
LookOut: Diverse Multi-Future Prediction and Planning for Self-Driving
Alexander Cui, Sergio Casas, Abbas Sadat, Renjie Liao, Raquel Urtasun
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 the diverse sampler (vs sampling from the prior) (ablation) | LookOut · ATG4D closed-loop simulation (ablation) | collision rate (%): 9.15 -> 7.93 | Improved |
| add the diverse sampler (vs sampling from the prior) (ablation) | LookOut · ATG4D closed-loop simulation (ablation) | progress (m): 64.90 -> 62.65 | No clear effect |
| add the planning diversity energy (ablation) | LookOut · ATG4D closed-loop simulation (ablation) | collision rate (%): 10.98 -> 7.93 | Improved |
| add the planning diversity energy (ablation) | LookOut · ATG4D closed-loop simulation (ablation) | lateral acceleration (m/s^2): 0.20 -> 0.37 | Hurt |
| add the scenario scorer (ablation) | LookOut · ATG4D closed-loop simulation (ablation) | progress per collision (m): 658.58 -> 790.37 | Improved |
| add the scenario scorer (ablation) | LookOut · ATG4D closed-loop simulation (ablation) | collision rate (%): 9.15 -> 7.93 | Improved |
| use contingency planner instead of PLT planner (ablation) | LookOut · ATG4D closed-loop simulation (ablation) | collision rate (%): 12.80 -> 7.93 | Improved |
| use contingency planner instead of PLT planner (ablation) | LookOut · ATG4D closed-loop simulation (ablation) | progress per collision (m): 436.29 -> 790.37 | Improved |
| pair baseline with contingency planner instead of PLT | CVAE-DPP · ATG4D closed-loop simulation (baselines with contingency planner) | collision rate (%): 17.07 -> 12.80 | Improved |
| pair baseline with contingency planner instead of PLT | CVAE-DLow · ATG4D closed-loop simulation (baselines with contingency planner) | collision rate (%): 14.63 -> 9.76 | Improved |
| pair baseline with contingency planner instead of PLT | ILVM · ATG4D closed-loop simulation (baselines with contingency planner) | collision rate (%): 10.98 -> 9.15 | Improved |
| pair baseline with contingency planner instead of PLT | CVAE · ATG4D closed-loop simulation (baselines with contingency planner) | collision rate (%): 8.54 -> 9.76 | Hurt |
Closest papers by meaning
- Future-Aware End-to-End Driving: Bidirectional Modeling of Trajectory Planning and Scene Evolution (NeurIPS 2025 · similarity 0.89)
- Driving in the Occupancy World: Vision-Centric 4D Occupancy Forecasting and Planning via World Models for Autonomous Driving (AAAI 2025 · similarity 0.87)
- Hybrid-Driving: An Autonomous Driving Decision Framework Integrating Large Language Models, Knowledge Graphs and Driving Rules (AAAI 2025 · similarity 0.87)
- DriveGPT: Scaling Autoregressive Behavior Models for Driving (ICML 2025 · similarity 0.87)
- DriveFuture: Future-Aware Latent World Models for Autonomous Driving (NeurIPS 2026 · similarity 0.87)
- Motion Forecasting in Continuous Driving (NeurIPS 2024 · 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.
