# CrashChat: A Multimodal Large Language Model for Multitask Traffic Crash Video Analysis

arXiv: https://arxiv.org/abs/2512.18878 · HTML: https://mlsanity.com/paper/2512.18878/

## Key facts

- Cited by 1 papers 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 17 of its 24 references that are in the map.
- First posted on arXiv 21 Dec 2025.

## Placed in

- Multimodal LLMs / vision-language models / [Driving VLMs and open-vocabulary detection](https://mlsanity.com/mllm/driving-vlms-and-open-vocabulary-detection/)

## Builds on (most cited first, 17 of 17)

- [VideoLLaMA 3: Frontier Multimodal Foundation Models for Image and Video Understanding](https://mlsanity.com/paper/2501.13106/) (2025)
- [Which Tasks Should Be Learned Together in Multi-task Learning?](https://mlsanity.com/paper/1905.07553/) (2019)
- [Unsupervised Traffic Accident Detection in First-Person Videos](https://mlsanity.com/paper/1903.00618/) (2019)
- [DADA-2000: Can Driving Accident be Predicted by Driver Attention? Analyzed by A Benchmark](https://mlsanity.com/paper/1904.12634/) (2019)
- [Localizing Anomalies from Weakly-Labeled Videos](https://mlsanity.com/paper/2008.08944/) (2020)
- [Abductive Ego-View Accident Video Understanding for Safe Driving Perception](https://mlsanity.com/paper/2403.00436/) (2024)
- [D2-City: A Large-Scale Dashcam Video Dataset of Diverse Traffic Scenarios](https://mlsanity.com/paper/1904.01975/) (2019)
- [Nexar Dashcam Collision Prediction Dataset and Challenge](https://mlsanity.com/paper/2503.03848/) (2025)
- [Mitigating Task Interference in Multi-Task Learning via Explicit Task Routing with Non-Learnable Primitives](https://mlsanity.com/paper/2308.02066/) (2023)
- [AccidentGPT: Accident Analysis and Prevention from V2X Environmental Perception with Multi-modal Large Model](https://mlsanity.com/paper/2312.13156/) (2023)
- [Cognitive Accident Prediction in Driving Scenes: A Multimodality Benchmark](https://mlsanity.com/paper/2212.09381/) (2022)
- [ScVLM: Enhancing Vision-Language Model for Safety-Critical Event Understanding](https://mlsanity.com/paper/2410.00982/) (2024)
- [Text-Driven Traffic Anomaly Detection with Temporal High-Frequency Modeling in Driving Videos](https://mlsanity.com/paper/2401.03522/) (2024)
- [SafePLUG: Empowering Multimodal LLMs with Pixel-Level Insight and Temporal Grounding for Traffic Accident Understanding](https://mlsanity.com/paper/2508.06763/) (2025)
- [Simplifying Traffic Anomaly Detection with Video Foundation Models](https://mlsanity.com/paper/2507.09338/) (2025)
- [iFinder: Structured Zero-Shot Vision-Based LLM Grounding for Dash-Cam Video Reasoning](https://mlsanity.com/paper/2509.19552/) (2025)
- [A Gaze Data-based Comparative Study to Build a Trustworthy Human-AI Collaboration in Crash Anticipation](https://mlsanity.com/paper/2108.01599/) (2021)

## Built on by (most cited first, 1 of 1)

- [OmniTraffic: A Controllable Generation Pipeline and Benchmark for Spatio-Temporal Traffic Reasoning](https://mlsanity.com/paper/2606.15749/) (2026)

## Abstract (arXiv)

Automating crash video analysis is essential to leverage the growing availability of driving video data for traffic safety research and accountability attribution in autonomous driving. Crash video analysis is a challenging multitask problem due to the complex spatiotemporal dynamics of crash events in video data and the diverse analytical requirements involved. It requires capabilities spanning crash recognition, temporal grounding, and high-level video understanding. Existing models, however, cannot perform all these tasks within a unified framework, and effective training strategies for such models remain underexplored. To fill these gaps, this paper proposes CrashChat, a multimodal large language model (MLLM) for multitask traffic crash analysis, built upon VideoLLaMA3. CrashChat acquires domain-specific knowledge through instruction fine-tuning and employs a novel multitask learning strategy based on task decoupling and grouping, which maximizes the benefit of joint learning within and across task groups while mitigating negative transfer. Numerical experiments on consolidated public datasets demonstrate that CrashChat consistently outperforms existing MLLMs across model scales and traditional vision-based methods, achieving state-of-the-art performance. It reaches near-perfect accuracy in crash recognition, a 176% improvement in crash localization, and a 40% improvement in the more challenging pre-crash localization. Compared to general MLLMs, it substantially enhances textual accuracy and content coverage in crash description and reasoning tasks, with 0.18-0.41 increases in BLEU scores and 0.18-0.42 increases in ROUGE scores. Beyond its strong performance, CrashChat is a convenient, end-to-end analytical tool ready for practical implementation. The dataset and implementation code for CrashChat are available at https://github.com/Liangkd/CrashChat.