# Are Vision LLMs Road-Ready? A Comprehensive Benchmark for Safety-Critical Driving Video Understanding

arXiv: https://arxiv.org/abs/2504.14526 · HTML: https://mlsanity.com/paper/2504.14526/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- Placed in VLMs for autonomous driving, one of 9 directions in End-to-end autonomous driving.
- Builds on 24 of its 50 references that are in the map.
- First posted on arXiv 20 Apr 2025.

## Placed in

- End-to-end autonomous driving / [VLMs for autonomous driving](https://mlsanity.com/driving/vlms-for-autonomous-driving/)

## Builds on (most cited first, 24 of 24)

- [Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution](https://mlsanity.com/paper/2409.12191/) (2024)
- [Lost in the Middle: How Language Models Use Long Contexts](https://mlsanity.com/paper/2307.03172/) (2023)
- [LLaVA-OneVision: Easy Visual Task Transfer](https://mlsanity.com/paper/2408.03326/) (2024)
- [Scalability in Perception for Autonomous Driving: Waymo Open Dataset](https://mlsanity.com/paper/1912.04838/) (2019)
- [MMBench: Is Your Multi-modal Model an All-around Player?](https://mlsanity.com/paper/2307.06281/) (2023)
- [Video-MME: The First-Ever Comprehensive Evaluation Benchmark of Multi-modal LLMs in Video Analysis](https://mlsanity.com/paper/2405.21075/) (2024)
- [MVBench: A Comprehensive Multi-modal Video Understanding Benchmark](https://mlsanity.com/paper/2311.17005/) (2023)
- [MiniCPM-V: A GPT-4V Level MLLM on Your Phone](https://mlsanity.com/paper/2408.01800/) (2024)
- [GPT-Driver: Learning to Drive with GPT](https://mlsanity.com/paper/2310.01415/) (2023)
- [PLLaVA : Parameter-free LLaVA Extension from Images to Videos for Video Dense Captioning](https://mlsanity.com/paper/2404.16994/) (2024)
- [ToolQA: A Dataset for LLM Question Answering with External Tools](https://mlsanity.com/paper/2306.13304/) (2023)
- [NuScenes-QA: A Multi-modal Visual Question Answering Benchmark for Autonomous Driving Scenario](https://mlsanity.com/paper/2305.14836/) (2023)
- [Talk2Car: Taking Control of Your Self-Driving Car](https://mlsanity.com/paper/1909.10838/) (2019)
- [DRAMA: Joint Risk Localization and Captioning in Driving](https://mlsanity.com/paper/2209.10767/) (2022)
- [Explainable Object-induced Action Decision for Autonomous Vehicles](https://mlsanity.com/paper/2003.09405/) (2020)
- [A Survey of Safety on Large Vision-Language Models: Attacks, Defenses and Evaluations](https://mlsanity.com/paper/2502.14881/) (2025)
- [Rank2Tell: A Multimodal Driving Dataset for Joint Importance Ranking and Reasoning](https://mlsanity.com/paper/2309.06597/) (2023)
- [DME-Driver: Integrating Human Decision Logic and 3D Scene Perception in Autonomous Driving](https://mlsanity.com/paper/2401.03641/) (2024)
- [Can LVLMs Obtain a Driver's License? A Benchmark Towards Reliable AGI for Autonomous Driving](https://mlsanity.com/paper/2409.02914/) (2024)
- [ScVLM: Enhancing Vision-Language Model for Safety-Critical Event Understanding](https://mlsanity.com/paper/2410.00982/) (2024)
- [Enhancing Traffic Safety with Parallel Dense Video Captioning for End-to-End Event Analysis](https://mlsanity.com/paper/2404.08229/) (2024)
- [Mastering Long-Tail Complexity on Graphs: Characterization, Learning, and Generalization](https://mlsanity.com/paper/2305.09938/) (2023)
- [Towards Heterogeneous Long-tailed Learning: Benchmarking, Metrics, and Toolbox](https://mlsanity.com/paper/2307.08235/) (2023)
- [Towards Reliable Rare Category Analysis on Graphs via Individual Calibration](https://mlsanity.com/paper/2307.09858/) (2023)

## Built on by (most cited first, 13 of 13)

- [Perception, Reason, Think, and Plan: A Survey on Large Multimodal Reasoning Models](https://mlsanity.com/paper/2505.04921/) (2025)
- [Foundation Models in Autonomous Driving: A Survey on Scenario Generation and Scenario Analysis](https://mlsanity.com/paper/2506.11526/) (2025)
- [STSBench: A Spatio-temporal Scenario Benchmark for Multi-modal Large Language Models in Autonomous Driving](https://mlsanity.com/paper/2506.06218/) (2025)
- [WaymoQA: A Multi-View Visual Question Answering Dataset for Safety-Critical Reasoning in Autonomous Driving](https://mlsanity.com/paper/2511.20022/) (2025)
- [nuReasoning: A Reasoning-Centric Dataset and Benchmark for Long-Tail Autonomous Driving](https://mlsanity.com/paper/2605.31572/) (2026)
- [NuRisk: A Visual Question Answering Dataset for Agent-Level Risk Assessment in Autonomous Driving](https://mlsanity.com/paper/2509.25944/) (2025)
- [Embodied4C: Measuring What Matters for Embodied Vision-Language Navigation](https://mlsanity.com/paper/2512.18028/) (2025)
- [A knowledge-augmented dataset of high-risk driving scenarios with LLM annotations for autonomous driving](https://mlsanity.com/paper/2607.07103/) (2026)
- [Reducing Text Bias in Synthetically Generated MCQAs for VLMs in Autonomous Driving](https://mlsanity.com/paper/2602.17677/) (2026)
- [RoadSceneBench: A Lightweight Benchmark for Mid-Level Road Scene Understanding](https://mlsanity.com/paper/2511.22466/) (2025)
- [TPS-Drive: Task-Guided Representation Purification for VLM-based Autonomous Driving](https://mlsanity.com/paper/2605.27038/) (2026)
- [Bi-NAS: Towards Effective and Personalized Explanation for Recommender Systems via Bi-Level Neural Architecture Search](https://mlsanity.com/paper/2607.01387/) (2026)
- [Vision-Language-Action Autonomous Driving Agent with Language-based Memory](https://mlsanity.com/paper/2609.38641/) (2026)

## Abstract (arXiv)

Vision Large Language Models (VLLMs) have demonstrated impressive capabilities in general visual tasks such as image captioning and visual question answering. However, their effectiveness in specialized, safety-critical domains like autonomous driving remains largely unexplored. Autonomous driving systems require sophisticated scene understanding in complex environments, yet existing multimodal benchmarks primarily focus on normal driving conditions, failing to adequately assess VLLMs' performance in safety-critical scenarios. To address this, we introduce DVBench, a pioneering benchmark designed to evaluate the performance of VLLMs in understanding safety-critical driving videos. Built around a hierarchical ability taxonomy that aligns with widely adopted frameworks for describing driving scenarios used in assessing highly automated driving systems, DVBench features 10,000 multiple-choice questions with human-annotated ground-truth answers, enabling a comprehensive evaluation of VLLMs' capabilities in perception and reasoning. Experiments on 14 SOTA VLLMs, ranging from 0.5B to 72B parameters, reveal significant performance gaps, with no model achieving over 40% accuracy, highlighting critical limitations in understanding complex driving scenarios. To probe adaptability, we fine-tuned selected models using domain-specific data from DVBench, achieving accuracy gains ranging from 5.24 to 10.94 percentage points, with relative improvements of up to 43.59%. This improvement underscores the necessity of targeted adaptation to bridge the gap between general-purpose VLLMs and mission-critical driving applications. DVBench establishes an essential evaluation framework and research roadmap for developing VLLMs that meet the safety and robustness requirements for real-world autonomous systems. We released the benchmark toolbox and the fine-tuned model at: https://github.com/tong-zeng/DVBench.git.