Research map / Vision-language and multimodal / Robot action and navigation
VLR-Driver: Large Vision-Language-Reasoning Models for Embodied Autonomous Driving
Fanjie Kong, Yitong Li, Weihuang Chen, Chen Min, Yizhe Li, Zhiqiang Gao, Haoyang Li, Zhongyu Guo, Hongbin Sun
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 spatiotemporal CoT reasoning guidance (ablation) | VLR-Driver · Bench2Drive (CARLA, 44 routes) | Driving Score: 57.17 -> 75.01 | Improved |
| add spatiotemporal CoT reasoning guidance (ablation) | VLR-Driver · Bench2Drive (CARLA, 44 routes) | Success Rate: 34.09 -> 50.00 | Improved |
| use the VLR model instead of a standard LLM module (ablation) | VLR-Driver · Bench2Drive (CARLA, 44 routes) | Driving Score: 46.39 -> 75.01 | Improved |
| use the VLR model instead of a standard LLM module (ablation) | VLR-Driver · Bench2Drive (CARLA, 44 routes) | Success Rate: 20.45 -> 50.00 | Improved |
| add VLR-Driver dataset reasoning data (ablation) | VLR-Driver · Bench2Drive (CARLA, 44 routes) | Driving Score: 52.85 -> 75.01 | Improved |
| add VLR-Driver dataset reasoning data (ablation) | VLR-Driver · Bench2Drive (CARLA, 44 routes) | Success Rate: 27.27 -> 50.00 | Improved |
| add Step-GRPO reinforcement learning after LoRA SFT (ablation) | VLR-Driver · Bench2Drive (CARLA, 44 routes) | Driving Score: 65.57 -> 75.01 | Improved |
| add Step-GRPO reinforcement learning after LoRA SFT (ablation) | VLR-Driver · Bench2Drive (CARLA, 44 routes) | Success Rate: 45.45 -> 50.00 | Improved |
| add Step-GRPO reinforcement learning after LoRA SFT (ablation) | VLR-Driver · Bench2Drive (CARLA, 44 routes) | Emergency Brake ability: 63.63 -> 72.72 | Improved |
| VLR-Driver versus best prior baseline | VLR-Driver · Bench2Drive (CARLA, 44 routes) | Driving Score / Success Rate: 63.81 -> 75.01 DS; 40.90 -> 50.00 SR | Improved |
Closest papers by meaning
- Embodied Scene Understanding for Vision Language Models via MetaVQA (CVPR 2025 · similarity 0.87)
- DriveSpatial: A Benchmark for Spatiotemporal Intelligence in VLMs for Autonomous Driving (NeurIPS 2026 · similarity 0.87)
- Drive-R1: Bridging Reasoning and Planning in VLMs for Autonomous Driving with Reinforcement Learning (AAAI 2026 · similarity 0.86)
- VLP: Vision Language Planning for Autonomous Driving (CVPR 2024 · similarity 0.86)
- MindDriver: Introducing Progressive Multimodal Reasoning for Autonomous Driving (CVPR 2026 · similarity 0.86)
- SGDrive: Scene-to-Goal Hierarchical World Cognition for Autonomous Driving (CVPR 2026 · 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.
