# AdaThinkDrive: Adaptive Thinking via Reinforcement Learning for Autonomous Driving

arXiv: https://arxiv.org/abs/2509.13769 · HTML: https://mlsanity.com/paper/2509.13769/

## Key facts

- Cited by 58 papers in the map (data through 14 Oct 2026).
- Placed in Chain-of-thought reasoning for driving, one of 17 directions in LLM reasoning.
- Builds on 38 of its 41 references that are in the map.
- First posted on arXiv 17 Sep 2025.

## Placed in

- LLM reasoning / [Chain-of-thought reasoning for driving](https://mlsanity.com/lr/chain-of-thought-reasoning-for-driving/)
- End-to-end autonomous driving / [Reasoning VLA and VLM driving policies](https://mlsanity.com/driving/reasoning-vla-and-vlm-driving-policies/)
- Deep reinforcement learning methods / [RL for autonomous driving and traffic](https://mlsanity.com/deeprl/rl-for-autonomous-driving-and-traffic/)

## Builds on (most cited first, 30 of 38)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models](https://mlsanity.com/paper/2504.10479/) (2025)
- [Planning-oriented Autonomous Driving](https://mlsanity.com/paper/2212.10156/) (2022)
- [Vision-Language Models for Vision Tasks: A Survey](https://mlsanity.com/paper/2304.00685/) (2023)
- [VAD: Vectorized Scene Representation for Efficient Autonomous Driving](https://mlsanity.com/paper/2303.12077/) (2023)
- [DriveLM: Driving with Graph Visual Question Answering](https://mlsanity.com/paper/2312.14150/) (2023)
- [End-to-end Autonomous Driving: Challenges and Frontiers](https://mlsanity.com/paper/2306.16927/) (2023)
- [DriveVLM: The Convergence of Autonomous Driving and Large Vision-Language Models](https://mlsanity.com/paper/2402.12289/) (2024)
- [TransFuser: Imitation with Transformer-Based Sensor Fusion for Autonomous Driving](https://mlsanity.com/paper/2205.15997/) (2022)
- [NAVSIM: Data-Driven Non-Reactive Autonomous Vehicle Simulation and Benchmarking](https://mlsanity.com/paper/2406.15349/) (2024)
- [DiffusionDrive: Truncated Diffusion Model for End-to-End Autonomous Driving](https://mlsanity.com/paper/2411.15139/) (2024)
- [NuScenes-QA: A Multi-modal Visual Question Answering Benchmark for Autonomous Driving Scenario](https://mlsanity.com/paper/2305.14836/) (2023)
- [Hydra-MDP: End-to-end Multimodal Planning with Multi-target Hydra-Distillation](https://mlsanity.com/paper/2406.06978/) (2024)
- [EMMA: End-to-End Multimodal Model for Autonomous Driving](https://mlsanity.com/paper/2410.23262/) (2024)
- [ReCogDrive: A Reinforced Cognitive Framework for End-to-End Autonomous Driving](https://mlsanity.com/paper/2506.08052/) (2025)
- [ThinkPrune: Pruning Long Chain-of-Thought of LLMs via Reinforcement Learning](https://mlsanity.com/paper/2504.01296/) (2025)
- [ORION: A Holistic End-to-End Autonomous Driving Framework by Vision-Language Instructed Action Generation](https://mlsanity.com/paper/2503.19755/) (2025)
- [Senna: Bridging Large Vision-Language Models and End-to-End Autonomous Driving](https://mlsanity.com/paper/2410.22313/) (2024)
- [End-to-End Driving with Online Trajectory Evaluation via BEV World Model](https://mlsanity.com/paper/2504.01941/) (2025)
- [AdaptThink: Reasoning Models Can Learn When to Think](https://mlsanity.com/paper/2505.13417/) (2025)
- [LingoQA: Visual Question Answering for Autonomous Driving](https://mlsanity.com/paper/2312.14115/) (2023)
- [OmniDrive: A Holistic Vision-Language Dataset for Autonomous Driving with Counterfactual Reasoning](https://mlsanity.com/paper/2504.04348/) (2025)
- [GoalFlow: Goal-Driven Flow Matching for Multimodal Trajectories Generation in End-to-End Autonomous Driving](https://mlsanity.com/paper/2503.05689/) (2025)
- [Thinkless: LLM Learns When to Think](https://mlsanity.com/paper/2505.13379/) (2025)
- [AlphaDrive: Unleashing the Power of VLMs in Autonomous Driving via Reinforcement Learning and Reasoning](https://mlsanity.com/paper/2503.07608/) (2025)
- [OpenEMMA: Open-Source Multimodal Model for End-to-End Autonomous Driving](https://mlsanity.com/paper/2412.15208/) (2024)
- [DRAMA: An Efficient End-to-end Motion Planner for Autonomous Driving with Mamba](https://mlsanity.com/paper/2408.03601/) (2024)
- [AdaCoT: Pareto-Optimal Adaptive Chain-of-Thought Triggering via Reinforcement Learning](https://mlsanity.com/paper/2505.11896/) (2025)
- [Concise Reasoning via Reinforcement Learning](https://mlsanity.com/paper/2504.05185/) (2025)
- [Holistic Autonomous Driving Understanding by Bird's-Eye-View Injected Multi-Modal Large Models](https://mlsanity.com/paper/2401.00988/) (2024)

## Built on by (most cited first, 30 of 58)

- [Alpamayo-R1: Bridging Reasoning and Action Prediction for Generalizable Autonomous Driving in the Long Tail](https://mlsanity.com/paper/2511.00088/) (2025)
- [A Survey on Efficient Vision-Language-Action Models](https://mlsanity.com/paper/2510.24795/) (2025)
- [Vision-Language-Action Models for Autonomous Driving: Past, Present, and Future](https://mlsanity.com/paper/2512.16760/) (2025)
- [LaST-VLA: Thinking in Latent Spatio-Temporal Space for Vision-Language-Action in Autonomous Driving](https://mlsanity.com/paper/2603.01928/) (2026)
- [MindDrive: A Vision-Language-Action Model for Autonomous Driving via Online Reinforcement Learning](https://mlsanity.com/paper/2512.13636/) (2025)
- [DriveVA: Video Action Models are Zero-Shot Drivers](https://mlsanity.com/paper/2604.04198/) (2026)
- [Counterfactual VLA: Self-Reflective Vision-Language-Action Model with Adaptive Reasoning](https://mlsanity.com/paper/2512.24426/) (2025)
- [VGGDrive: Empowering Vision-Language Models with Cross-View Geometric Grounding for Autonomous Driving](https://mlsanity.com/paper/2602.20794/) (2026)
- [DriveFine: Refining-Augmented Masked Diffusion VLA for Precise and Robust Driving](https://mlsanity.com/paper/2602.14577/) (2026)
- [Unleashing VLA Potentials in Autonomous Driving via Explicit Learning from Failures](https://mlsanity.com/paper/2603.01063/) (2026)
- [UniDriveVLA: Unifying Understanding, Perception, and Action Planning for Autonomous Driving](https://mlsanity.com/paper/2604.02190/) (2026)
- [NoRD: A Data-Efficient Vision-Language-Action Model that Drives without Reasoning](https://mlsanity.com/paper/2602.21172/) (2026)
- [Devil is in Narrow Policy: Unleashing Exploration in Driving VLA Models](https://mlsanity.com/paper/2603.06049/) (2026)
- [dVLM-AD: Enhance Diffusion Vision-Language-Model for Driving via Controllable Reasoning](https://mlsanity.com/paper/2512.04459/) (2025)
- [DVGT-2: Vision-Geometry-Action Model for Autonomous Driving at Scale](https://mlsanity.com/paper/2604.00813/) (2026)
- [CoWorld-VLA: Thinking in a Multi-Expert World Model for Autonomous Driving](https://mlsanity.com/paper/2605.10426/) (2026)
- [DynVLA: Learning World Dynamics for Action Reasoning in Autonomous Driving](https://mlsanity.com/paper/2603.11041/) (2026)
- [Metis: A Generalizable and Efficient World-Action Model for Autonomous Driving and Urban Navigation](https://mlsanity.com/paper/2606.15869/) (2026)
- [MindDriver: Introducing Progressive Multimodal Reasoning for Autonomous Driving](https://mlsanity.com/paper/2602.21952/) (2026)
- [Bench2Drive-VL: Benchmarks for Closed-Loop Autonomous Driving with Vision-Language Models](https://mlsanity.com/paper/2604.01259/) (2026)
- [Xiaomi OneVL: One-Step Latent Reasoning and Planning with Vision-Language Explanation](https://mlsanity.com/paper/2604.18486/) (2026)
- [MindVLA-U1: VLA Beats VA with Unified Streaming Architecture for Autonomous Driving](https://mlsanity.com/paper/2605.12624/) (2026)
- [Auto-JEPA: A Latent World Model of Continuous Intent for End-to-End Autonomous Driving](https://mlsanity.com/paper/2607.29031/) (2026)
- [Distill to Think, Foresee to Act: Cognitive-Physical Reinforcement Learning for Autonomous Driving](https://mlsanity.com/paper/2605.21139/) (2026)
- [VLGA: Vision-Language-Geometry-Action Models for Autonomous Driving](https://mlsanity.com/paper/2606.12396/) (2026)
- [IDOL: Inverse-Dynamics-Guided Future Prediction for End-to-End Autonomous Driving](https://mlsanity.com/paper/2605.31476/) (2026)
- [DriveReward: A Comprehensive Dataset and Generative Vision-Language Reward Model for Autonomous Driving](https://mlsanity.com/paper/2606.08525/) (2026)
- [Intend, Reflect, Refine: An Adaptive Multimodal Reflection Framework for Autonomous Driving](https://mlsanity.com/paper/2606.22913/) (2026)
- [Post-Training in End-to-End Autonomous Driving](https://mlsanity.com/paper/2607.08072/) (2026)
- [GeoWorldAD: Geometry World Action Model for Autonomous Driving](https://mlsanity.com/paper/2607.17521/) (2026)

## Abstract (arXiv)

While reasoning technology like Chain of Thought (CoT) has been widely adopted in Vision Language Action (VLA) models, it demonstrates promising capabilities in end to end autonomous driving. However, recent efforts to integrate CoT reasoning often fall short in simple scenarios, introducing unnecessary computational overhead without improving decision quality. To address this, we propose AdaThinkDrive, a novel VLA framework with a dual mode reasoning mechanism inspired by fast and slow thinking. First, our framework is pretrained on large scale autonomous driving (AD) scenarios using both question answering (QA) and trajectory datasets to acquire world knowledge and driving commonsense. During supervised fine tuning (SFT), we introduce a two mode dataset, fast answering (w/o CoT) and slow thinking (with CoT), enabling the model to distinguish between scenarios that require reasoning. Furthermore, an Adaptive Think Reward strategy is proposed in conjunction with the Group Relative Policy Optimization (GRPO), which rewards the model for selectively applying CoT by comparing trajectory quality across different reasoning modes. Extensive experiments on the Navsim benchmark show that AdaThinkDrive achieves a PDMS of 90.3, surpassing the best vision only baseline by 1.7 points. Moreover, ablations show that AdaThinkDrive surpasses both the never Think and always Think baselines, improving PDMS by 2.0 and 1.4, respectively. It also reduces inference time by 14% compared to the always Think baseline, demonstrating its ability to balance accuracy and efficiency through adaptive reasoning.