TAU-R1: Visual Language Model for Traffic Anomaly Understanding
This work addresses traffic safety in Intelligent Transportation Systems by providing a benchmark and method for traffic anomaly understanding, though it is incremental as it builds on existing vision-language models.
The authors tackled the problem of Traffic Anomaly Understanding (TAU) by introducing Roundabout-TAU, a dataset with 342 clips and over 2,000 question-answer pairs, and TAU-R1, a two-layer vision-language framework that achieves strong performance on anomaly classification and reasoning tasks while maintaining deployment efficiency.
Traffic Anomaly Understanding (TAU) is important for traffic safety in Intelligent Transportation Systems. Recent vision-language models (VLMs) have shown strong capabilities in video understanding. However, progress on TAU remains limited due to the lack of benchmarks and task-specific methodologies. To address this limitation, we introduce Roundabout-TAU, a dataset constructed from real-world roundabout videos collected in collaboration with the City of Carmel, Indiana. The dataset contains 342 clips and is annotated with more than 2,000 question-answer pairs covering multiple aspects of traffic anomaly understanding. Building on this benchmark, we propose TAU-R1, a two-layer vision-language framework for TAU. The first layer is a lightweight anomaly classifier that performs coarse anomaly categorisation, while the second layer is a larger anomaly reasoner that generates detailed event summaries. To improve task-specific reasoning, we introduce a two-stage training strategy consisting of decomposed-QA-enhanced supervised fine-tuning followed by TAU-GRPO, a GRPO-based post-training method with TAU-specific reward functions. Experimental results show that TAU-R1 achieves strong performance on both anomaly classification and reasoning tasks while maintaining deployment efficiency. The dataset and code are available at: https://github.com/siri-rouser/TAU-R1