# VERDI: VLM-Embedded Reasoning for Autonomous Driving

arXiv: https://arxiv.org/abs/2505.15925 · HTML: https://mlsanity.com/paper/2505.15925/

## Key facts

- Cited by 16 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 47 of its 57 references that are in the map.
- First posted on arXiv 21 May 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/)

## Builds on (most cited first, 30 of 47)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [nuScenes: A multimodal dataset for autonomous driving](https://mlsanity.com/paper/1903.11027/) (2019)
- [Supervised Contrastive Learning](https://mlsanity.com/paper/2004.11362/) (2020)
- [Scalability in Perception for Autonomous Driving: Waymo Open Dataset](https://mlsanity.com/paper/1912.04838/) (2019)
- [CARLA: An Open Urban Driving Simulator](https://mlsanity.com/paper/1711.03938/) (2017)
- [Knowledge Distillation: A Survey](https://mlsanity.com/paper/2006.05525/) (2020)
- [Reproducible scaling laws for contrastive language-image learning](https://mlsanity.com/paper/2212.07143/) (2022)
- [Planning-oriented Autonomous Driving](https://mlsanity.com/paper/2212.10156/) (2022)
- [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)
- [DriveGPT4: Interpretable End-to-end Autonomous Driving via Large Language Model](https://mlsanity.com/paper/2310.01412/) (2023)
- [ST-P3: End-to-end Vision-based Autonomous Driving via Spatial-Temporal Feature Learning](https://mlsanity.com/paper/2207.07601/) (2022)
- [VADv2: End-to-End Vectorized Autonomous Driving via Probabilistic Planning](https://mlsanity.com/paper/2402.13243/) (2024)
- [A Survey on Knowledge Distillation of Large Language Models](https://mlsanity.com/paper/2402.13116/) (2024)
- [Bench2Drive: Towards Multi-Ability Benchmarking of Closed-Loop End-To-End Autonomous Driving](https://mlsanity.com/paper/2406.03877/) (2024)
- [Trajectory-guided Control Prediction for End-to-end Autonomous Driving: A Simple yet Strong Baseline](https://mlsanity.com/paper/2206.08129/) (2022)
- [End-to-end Interpretable Neural Motion Planner](https://mlsanity.com/paper/2101.06679/) (2021)
- [Is Ego Status All You Need for Open-Loop End-to-End Autonomous Driving?](https://mlsanity.com/paper/2312.03031/) (2023)
- [DriveMLM: Aligning Multi-Modal Large Language Models with Behavioral Planning States for Autonomous Driving](https://mlsanity.com/paper/2312.09245/) (2023)
- [Driving with LLMs: Fusing Object-Level Vector Modality for Explainable Autonomous Driving](https://mlsanity.com/paper/2310.01957/) (2023)
- [EMMA: End-to-End Multimodal Model for Autonomous Driving](https://mlsanity.com/paper/2410.23262/) (2024)
- [Safety-Enhanced Autonomous Driving Using Interpretable Sensor Fusion Transformer](https://mlsanity.com/paper/2207.14024/) (2022)
- [BEVerse: Unified Perception and Prediction in Birds-Eye-View for Vision-Centric Autonomous Driving](https://mlsanity.com/paper/2205.09743/) (2022)
- [Think Twice before Driving: Towards Scalable Decoders for End-to-End Autonomous Driving](https://mlsanity.com/paper/2305.06242/) (2023)

## Built on by (most cited first, 16 of 16)

- [Vision-Language-Action Models for Autonomous Driving: Past, Present, and Future](https://mlsanity.com/paper/2512.16760/) (2025)
- [WaymoQA: A Multi-View Visual Question Answering Dataset for Safety-Critical Reasoning in Autonomous Driving](https://mlsanity.com/paper/2511.20022/) (2025)
- [LLaViDA: A Large Language Vision Driving Assistant for Explicit Reasoning and Enhanced Trajectory Planning](https://mlsanity.com/paper/2512.18211/) (2025)
- [NaviDriveVLM: Decoupling High-Level Reasoning and Motion Planning for Autonomous Driving](https://mlsanity.com/paper/2603.07901/) (2026)
- [SEAL: Vision-Language Model-Based Safe End-to-End Cooperative Autonomous Driving with Adaptive Long-Tail Modeling](https://mlsanity.com/paper/2506.21041/) (2025)
- [Latent World Models for Automated Driving: A Unified Taxonomy, Evaluation Framework, and Open Challenges](https://mlsanity.com/paper/2603.09086/) (2026)
- [How Well Do Vision-Language Models Understand Sequential Driving Scenes? A Sensitivity Study](https://mlsanity.com/paper/2604.06750/) (2026)
- [OpenLongTail: Generative Scaling of Long-Tail Driving Data](https://mlsanity.com/paper/2607.09655/) (2026)
- [Less is More: Lean yet Powerful Vision-Language Model for Autonomous Driving](https://mlsanity.com/paper/2510.00060/) (2025)
- [BEVLM: Distilling Semantic Knowledge from LLMs into Bird's-Eye View Representations](https://mlsanity.com/paper/2603.06576/) (2026)
- [Orion-Lite: Distilling LLM Reasoning into Efficient Vision-Only Driving Models](https://mlsanity.com/paper/2604.08266/) (2026)
- [Cross-Stage Coherence in Hierarchical Driving VQA: Explicit Baselines and Learned Gated Context Projectors](https://mlsanity.com/paper/2604.22560/) (2026)
- [Multi-Agent Orchestration with the Common-Sense Reasoning Capabilities of LLMs for Autonomous Driving](https://mlsanity.com/paper/2608.20129/) (2026)
- [Beyond Textual Chain-of-Thought: A Survey on Action-Grounded Reasoning in Autonomous Driving](https://mlsanity.com/paper/2609.01659/) (2026)
- [CrashDiffuser: VLM-Guided Collision Intent Reasoning for Fine-Grained Safety-Critical Traffic Scenario Generation](https://mlsanity.com/paper/2609.02270/) (2026)
- [Training Generalist Navigation Policies via Action Self-Distillation](https://mlsanity.com/paper/2610.17593/) (2026)

## Abstract (arXiv)

While autonomous driving (AD) stacks struggle with decision making under partial observability and real-world complexity, human drivers are capable of applying commonsense reasoning to make near-optimal decisions with limited information. Recent work has attempted to leverage finetuned Vision-Language Models (VLMs) for trajectory planning at inference time to emulate human behavior. Despite their success in benchmark evaluations, these methods are often impractical to deploy (a 70B parameter VLM inference at merely 8 tokens per second requires more than 160G of memory), and their monolithic network structure prohibits safety decomposition. To bridge this gap, we propose VLM-Embedded Reasoning for autonomous DrIving (VERDI), a training-time framework that distills the reasoning process and commonsense knowledge of VLMs into the AD stack. VERDI augments modular differentiable end-to-end (e2e) AD models by aligning intermediate module outputs at the perception, prediction, and planning stages with text features explaining the driving reasoning process produced by VLMs. By encouraging alignment in latent space, VERDI enables the modular AD stack to internalize structured reasoning, without incurring the inference-time costs of large VLMs. We evaluate VERDI in both open-loop and closed-loop settings. Our method outperforms existing end-to-end approaches without embedded reasoning by up to 11% in ℓ2 distance, and achieves the best overall driving performance in the closed-loop HugSim simulator, including a 10% improvement in Non-Collision Rate, while maintaining fast inference speed.