# Enhancing Traffic Incident Response through Sub-Second Temporal Localization with HybridMamba

arXiv: https://arxiv.org/abs/2504.03235 · HTML: https://mlsanity.com/paper/2504.03235/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 26 of its 45 references that are in the map.
- First posted on arXiv 4 Apr 2025.

## Builds on (most cited first, 26 of 26)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [Mamba: Linear-Time Sequence Modeling with Selective State Spaces](https://mlsanity.com/paper/2312.00752/) (2023)
- [SlowFast Networks for Video Recognition](https://mlsanity.com/paper/1812.03982/) (2018)
- [SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features](https://mlsanity.com/paper/2502.14786/) (2025)
- [Video Swin Transformer](https://mlsanity.com/paper/2106.13230/) (2021)
- [Video-ChatGPT: Towards Detailed Video Understanding via Large Vision and Language Models](https://mlsanity.com/paper/2306.05424/) (2023)
- [CLIP-Adapter: Better Vision-Language Models with Feature Adapters](https://mlsanity.com/paper/2110.04544/) (2021)
- [VideoChat: Chat-Centric Video Understanding](https://mlsanity.com/paper/2305.06355/) (2023)
- [VideoLLaMA 2: Advancing Spatial-Temporal Modeling and Audio Understanding in Video-LLMs](https://mlsanity.com/paper/2406.07476/) (2024)
- [Otter: A Multi-Modal Model with In-Context Instruction Tuning](https://mlsanity.com/paper/2305.03726/) (2023)
- [TimeChat: A Time-sensitive Multimodal Large Language Model for Long Video Understanding](https://mlsanity.com/paper/2312.02051/) (2023)
- [A Simple LLM Framework for Long-Range Video Question-Answering](https://mlsanity.com/paper/2312.17235/) (2023)
- [Macaw-LLM: Multi-Modal Language Modeling with Image, Audio, Video, and Text Integration](https://mlsanity.com/paper/2306.09093/) (2023)
- [VideoDirectorGPT: Consistent Multi-scene Video Generation via LLM-Guided Planning](https://mlsanity.com/paper/2309.15091/) (2023)
- [X-LLM: Bootstrapping Advanced Large Language Models by Treating Multi-Modalities as Foreign Languages](https://mlsanity.com/paper/2305.04160/) (2023)
- [LITA: Language Instructed Temporal-Localization Assistant](https://mlsanity.com/paper/2403.19046/) (2024)
- [LLM-grounded Video Diffusion Models](https://mlsanity.com/paper/2309.17444/) (2023)
- [TimeMarker: A Versatile Video-LLM for Long and Short Video Understanding with Superior Temporal Localization Ability](https://mlsanity.com/paper/2411.18211/) (2024)
- [Predicting Driver Attention in Critical Situations](https://mlsanity.com/paper/1711.06406/) (2017)
- [CADP: A Novel Dataset for CCTV Traffic Camera based Accident Analysis](https://mlsanity.com/paper/1809.05782/) (2018)
- [D2-City: A Large-Scale Dashcam Video Dataset of Diverse Traffic Scenarios](https://mlsanity.com/paper/1904.01975/) (2019)
- [inTformer: A Time-Embedded Attention-Based Transformer for Crash Likelihood Prediction at Intersections Using Connected Vehicle Data](https://mlsanity.com/paper/2307.03854/) (2023)

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

- [Image Segmentation with Large Language Models: A Survey with Perspectives for Intelligent Transportation Systems](https://mlsanity.com/paper/2506.14096/) (2025)

## Abstract (arXiv)

Traffic crash detection in long-form surveillance videos is essential for improving emergency response and infrastructure planning, yet remains difficult due to the brief and infrequent nature of crash events. We present HybridMamba, a novel architecture integrating visual transformers with state-space temporal modeling to achieve high-precision crash time localization. Our approach introduces multi-level token compression and hierarchical temporal processing to maintain computational efficiency without sacrificing temporal resolution. Evaluated on a large-scale dataset from the Iowa Department of Transportation, HybridMamba achieves a mean absolute error of 1.50 seconds for 2-minute videos (p<0.01 compared to baselines), with 65.2% of predictions falling within one second of the ground truth. It outperforms recent video-language models (e.g., TimeChat, VideoLLaMA-2) by up to 3.95 seconds while using significantly fewer parameters (3B vs. 13--72B). Our results demonstrate effective temporal localization across various video durations (2--40 minutes) and diverse environmental conditions, highlighting HybridMamba's potential for fine-grained temporal localization in traffic surveillance while identifying challenges that remain for extended deployment.