Research map / Large language models / Agents, code and math
AdaDrive: Self-Adaptive Slow-Fast System for Language-Grounded Autonomous Driving
Ruifei Zhang, Junlin Xie, Wei Zhang, Weikai Chen, Xiao Tan, Xiang Wan, Guanbin 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 |
|---|---|---|---|
| add LS-Qformer in place of vanilla Q-former (ablation) | planner baseline · LangAuto-Tiny | driving score: 67.4 -> 71.9 | Improved |
| add Connector-W adaptive LLM activation (ablation) | AdaDrive with LS-Qformer · LangAuto-Tiny | driving score: 71.9 -> 77.9 | Improved |
| add Connector-H confidence-scaled fusion (ablation) | AdaDrive with Connector-W · LangAuto-Tiny | driving score: 77.9 -> 80.9 | Improved |
| add Connector-H confidence-scaled fusion (ablation) | AdaDrive with Connector-W · LangAuto-Tiny | infraction score: 0.89 -> 0.90 | Improved |
| use LS-Qformer instead of vanilla Q-former (ablation) | AdaDrive · LangAuto-Tiny | driving score: 75.8 -> 80.9 | Improved |
| use LS-Qformer instead of SeqQ-Former (ablation) | AdaDrive · LangAuto-Tiny | driving score: 77.6 -> 80.9 | Improved |
| use LS-Qformer instead of additive history Q-former (ablation) | AdaDrive · LangAuto-Tiny | driving score: 77.4 -> 80.9 | Improved |
| use adaptive activation instead of fixed-interval activation at similar frequency (0.25) (ablation) | AdaDrive · LangAuto-Short | driving score: 5.6% driving score gain | Improved |
| use adaptive activation instead of continuous activation (ablation) | AdaDrive · LangAuto-Short | GFLOPs: reducing GFLOPs by 62% | Improved |
| use propagative memory fusion instead of hard update buffer (ablation) | AdaDrive · LangAuto-Short | driving score | Improved |
| remove streaming memory buffer (current frame only) (ablation) | AdaDrive · LangAuto-Short | driving score | No clear effect |
| use larger memory buffer capacity (ablation) | AdaDrive · LangAuto-Short | driving score | Hurt |
Closest papers by meaning
- LaMPilot: An Open Benchmark Dataset for Autonomous Driving with Language Model Programs (CVPR 2024 · similarity 0.90)
- Hybrid-Driving: An Autonomous Driving Decision Framework Integrating Large Language Models, Knowledge Graphs and Driving Rules (AAAI 2025 · similarity 0.90)
- Driving Everywhere with Large Language Model Policy Adaptation (CVPR 2024 · similarity 0.89)
- AdaWM: Adaptive World Model based Planning for Autonomous Driving (ICLR 2025 · similarity 0.88)
- HCRMP: An LLM-Hinted Contextual Reinforcement Learning Framework for Autonomous Driving (NeurIPS 2025 · similarity 0.88)
- AutoVLA: A Vision-Language-Action Model for End-to-End Autonomous Driving with Adaptive Reasoning and Reinforcement Fine-Tuning (NeurIPS 2025 · similarity 0.88)
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.
