mlsanityData to 14 Oct 2026

HazardNet: A Small-Scale Vision Language Model for Real-Time Traffic Safety Detection at Edge Devices

arXiv 2502.20572 ↗First posted 27 Feb 2025cs.CV cs.CLcited by 1

On the map: Multimodal LLMs / vision-language models → Driving VLMs and open-vocabulary detection

Abstract (arXiv)

Traffic safety remains a vital concern in contemporary urban settings, intensified by the increase of vehicles and the complicated nature of road networks. Traditional safety-critical event detection systems predominantly rely on sensor-based approaches and conventional machine learning algorithms, necessitating extensive data collection and complex training processes to adhere to traffic safety regulations. This paper introduces HazardNet, a small-scale Vision Language Model designed to enhance traffic safety by leveraging the reasoning capabilities of advanced language and vision models. We built HazardNet by fine-tuning the pre-trained Qwen2-VL-2B model, chosen for its superior performance among open-source alternatives and its compact size of two billion parameters. This helps to facilitate deployment on edge devices with efficient inference throughput. In addition, we present HazardQA, a novel Vision Question Answering (VQA) dataset constructed specifically for training HazardNet on real-world scenarios involving safety-critical events. Our experimental results show that the fine-tuned HazardNet outperformed the base model up to an 89% improvement in F1-Score and has comparable results with improvement in some cases reach up to 6% when compared to larger models, such as GPT-4o. These advancements underscore the potential of HazardNet in providing real-time, reliable traffic safety event detection, thereby contributing to reduced accidents and improved traffic management in urban environments. Both HazardNet model and the HazardQA dataset are available at https://huggingface.co/Tami3/HazardNet and https://huggingface.co/datasets/Tami3/HazardQA, respectively.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Placed in Driving VLMs and open-vocabulary detection, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 20 of its 31 references that are in the map. First posted on arXiv 27 Feb 2025.

Builds on

20 of 31 references.

Closest ancestors · 4

Same direction, newest first

  1. 2024Advancing Object Detection in Transportation with Multimodal Large Language Models (MLLMs): A Comprehensive Review and Empirical Testing26 Sep
  2. 2023DriveLM: Driving with Graph Visual Question Answering21 Dec
  3. 2023DriveMLM: Aligning Multi-Modal Large Language Models with Behavioral Planning States for Autonomous Driving14 Dec
  4. 2023A Survey on Multimodal Large Language Models for Autonomous Driving21 Nov

Other references · 16

  1. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,380
  2. 2019nuScenes: A multimodal dataset for autonomous drivingcited by 4,523
  3. 2024Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolutioncited by 4,234
  4. 2023QLoRA: Efficient Finetuning of Quantized LLMscited by 3,418
  5. 2019Scalability in Perception for Autonomous Driving: Waymo Open Datasetcited by 2,451
  6. 2023Vision-Language Models for Vision Tasks: A Surveycited by 839
  7. 2023GPT-Driver: Learning to Drive with GPTcited by 347
  8. 2023Driving with LLMs: Fusing Object-Level Vector Modality for Explainable Autonomous Drivingcited by 232
  9. 2023LanguageMPC: Large Language Models as Decision Makers for Autonomous Drivingcited by 190
  10. 2023LLM4Drive: A Survey of Large Language Models for Autonomous Drivingcited by 136
  11. 2022DRAMA: Joint Risk Localization and Captioning in Drivingcited by 126
  12. 2022KING: Generating Safety-Critical Driving Scenarios for Robust Imitation via Kinematics Gradientscited by 96
4 more
  1. 2024Visual Reasoning and Multi-Agent Approach in Multimodal Large Language Models (MLLMs): Solving TSP and mTSP Combinatorial Challengescited by 21
  2. 2024Large Language Models (LLMs) as Traffic Control Systems at Urban Intersections: A New Paradigmcited by 12
  3. 2023Categorical Traffic Transformer: Interpretable and Diverse Behavior Prediction with Tokenized Latentcited by 10
  4. 2024Leveraging Large Language Models (LLMs) for Traffic Management at Urban Intersections: The Case of Mixed Traffic Scenarioscited by 6

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026EG-ARSA: An Expert-Grounded Open Model for Visual Road Safety Auditing in Low-Resource Settingscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025INSIGHT: Enhancing Autonomous Driving Safety through Vision-Language Models on Context-Aware Hazard Detection and Edge Case Evaluation
  2. 2025Towards a Multi-Agent Vision-Language System for Zero-Shot Novel Hazardous Object Detection for Autonomous Driving Safety
  3. 2026Urban Risk-Aware Navigation via VQA-Based Event Maps for People with Low Vision
  4. 2026AUTOPILOT VQA: Benchmarking Vision-Language Models for Incident-Centric Dashcam Understanding
  5. 2025Multimodal Large Language Models for Enhanced Traffic Safety: A Comprehensive Review and Future Trends
  6. 2025SafePLUG: Empowering Multimodal LLMs with Pixel-Level Insight and Temporal Grounding for Traffic Accident Understanding
  7. 2025VRU-Accident: A Vision-Language Benchmark for Video Question Answering and Dense Captioning for Accident Scene Understanding
  8. 2025When language and vision meet road safety: leveraging multimodal large language models for video-based traffic accident analysis