Research map / Detection and segmentation / CNN features
ReasonNet: End-to-End Driving With Temporal and Global Reasoning
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 |
|---|---|---|---|
| Add temporal memory buffers for historical scene reasoning (ablation) | ReasonNet · Town05 Long | DS: 66.7±3.8 -> 73.2±1.9 | Helped |
| Add temporal memory buffers for tracking intermittently occluded objects (ablation) | ReasonNet · DOS | SR#1: 22±1.6 -> 63±4.2 | Helped |
| Add long-term memory buffer (ablation) | ReasonNet with Ts=4 · DOS | SR#3: 40±5.5 -> 80±4.2 | Helped |
| Select long-term memory using both object probability and usage frequency (ablation) | ReasonNet · DOS | SR#1: 33±4.4 -> 63±4.2 | Helped |
| Select long-term memory using object probability rather than usage frequency alone (ablation) | ReasonNet · DOS | SR#1: 45±4.2 -> 57±4.1 | Helped |
| Add global reasoning over object-environment and object-object interactions (ablation) | ReasonNet · DOS | SR#1: 28±2.8 -> 63±4.2 | Helped |
| Add the consistency loss between waypoints and occupancy predictions (ablation) | ReasonNet · Town05 Long | DS: 72.2±3.4 -> 73.2±1.9 | Helped |
| Add global-reasoning traffic sign features to the final traffic sign prediction (ablation) | ReasonNet · Town05 Long | Red: 0.05±0.03 -> 0.03±0.01 | Helped |
| Use ReasonNet instead of InterFuser | CARLA Leaderboard · CARLA Leaderboard | DS: 76.18 -> 79.95 | Helped |
| Use ReasonNet instead of InterFuser | Town05 Long · Town05 Long | Driving Score: 68.31 ± 1.86 -> 73.22 ± 1.91 | Helped |
Closest papers by meaning
- nuReasoning: A Reasoning-Centric Dataset and Benchmark for Long-Tail Autonomous Driving (NeurIPS 2026 · similarity 0.89)
- DriveCombo: Benchmarking Compositional Traffic Rule Reasoning in Autonomous Driving (CVPR 2026 · similarity 0.89)
- DriveSpatial: A Benchmark for Spatiotemporal Intelligence in VLMs for Autonomous Driving (NeurIPS 2026 · similarity 0.88)
- STSBench: A Spatio-temporal Scenario Benchmark for Multi-modal Large Language Models in Autonomous Driving (NeurIPS 2025 · similarity 0.88)
- Enhancing End-to-End Autonomous Driving with Latent World Model (ICLR 2025 · similarity 0.88)
- Future-Aware End-to-End Driving: Bidirectional Modeling of Trajectory Planning and Scene Evolution (NeurIPS 2025 · 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.