# Drive by Hindsight and Foresight: Tool-Grounded Synergistic Reasoning over Hierarchical Memory for Autonomous Driving

arXiv: https://arxiv.org/abs/2609.08217 · HTML: https://mlsanity.com/paper/2609.08217/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Reasoning VLA and VLM driving policies, one of 9 directions in End-to-end autonomous driving.
- Builds on 13 of its 33 references that are in the map.
- First posted on arXiv 8 Sep 2026.

## Placed in

- End-to-end autonomous driving / [Reasoning VLA and VLM driving policies](https://mlsanity.com/driving/reasoning-vla-and-vlm-driving-policies/)

## Builds on (most cited first, 13 of 13)

- [NuScenes-QA: A Multi-modal Visual Question Answering Benchmark for Autonomous Driving Scenario](https://mlsanity.com/paper/2305.14836/) (2023)
- [EMMA: End-to-End Multimodal Model for Autonomous Driving](https://mlsanity.com/paper/2410.23262/) (2024)
- [ORION: A Holistic End-to-End Autonomous Driving Framework by Vision-Language Instructed Action Generation](https://mlsanity.com/paper/2503.19755/) (2025)
- [A Language Agent for Autonomous Driving](https://mlsanity.com/paper/2311.10813/) (2023)
- [Alpamayo-R1: Bridging Reasoning and Action Prediction for Generalizable Autonomous Driving in the Long Tail](https://mlsanity.com/paper/2511.00088/) (2025)
- [LingoQA: Visual Question Answering for Autonomous Driving](https://mlsanity.com/paper/2312.14115/) (2023)
- [Are VLMs Ready for Autonomous Driving? An Empirical Study from the Reliability, Data, and Metric Perspectives](https://mlsanity.com/paper/2501.04003/) (2025)
- [Spatial Reasoning with Vision-Language Models in Ego-Centric Multi-View Scenes](https://mlsanity.com/paper/2509.06266/) (2025)
- [WaymoQA: A Multi-View Visual Question Answering Dataset for Safety-Critical Reasoning in Autonomous Driving](https://mlsanity.com/paper/2511.20022/) (2025)
- [OmniDrive-R1: Reinforcement-driven Interleaved Multi-modal Chain-of-Thought for Trustworthy Vision-Language Autonomous Driving](https://mlsanity.com/paper/2512.14044/) (2025)
- [NuRisk: A Visual Question Answering Dataset for Agent-Level Risk Assessment in Autonomous Driving](https://mlsanity.com/paper/2509.25944/) (2025)
- [Judge, Then Drive: A Critic-Centric Vision Language Action Framework for Autonomous Driving](https://mlsanity.com/paper/2604.27366/) (2026)
- [CritiqueDriveVLM: From Verifier-Guided Reinforcement Learning to Latent Thought Distillation for Autonomous Driving](https://mlsanity.com/paper/2607.04179/) (2026)

## Abstract (arXiv)

VLMs have shown promise for autonomous driving, yet still suffer from hallucination, weak spatio-temporal perception, and limited generalization. Recent methods improve reasoning and decision-making through CoT explanations, retrieval-augmented generation or the static injection of tool outputs. Although these mechanisms enrich the context, the model neither proactively perceives scene information nor accumulates experience after answering. To overcome these limitations, we present, to our knowledge, the first synergistic framework that tightly couples hierarchical memory with proactive tool invocation in a closed reasoning loop. Our contributions are threefold. (i) Hierarchical Driving Memory: a scene-level short-term memory maintains the dynamic scene state, and an evolving long-term memory retrieves reusable experience and tool strategies. (ii) Memory-Tool Synergistic Reasoning Framework: guided by the scene state and retrieved experience, the model adaptively invokes tools to refine its reasoning at inference time and consolidates reusable experience into a long-term memory pool offline. (iii) Data Generation and Two-stage Training Pipeline: verified memory-tool trajectories built by multi-step teacher rollout are used to train with SFT and GRPO. Our 7B model reaches an overall reasoning score of 80.03 and MCQ accuracy of 79.09% on DriveLMM-o1, surpassing the strongest baseline by 7.74 MCQ points and generalizes strongly across benchmarks. Notably, ablation and analysis studies validate the effectiveness of each component and further reveal the complementary roles of hierarchical memory. Short-term memory strengthens spatio-temporal understanding, improving STSBench accuracy by 24.2 points, while offline long-term memory consolidation yields an additional 3.57-point MCQ gain with all parameters frozen, demonstrating continual self-evolution through accumulated driving experience.