# Efficient Driving Behavior Narration and Reasoning on Edge Device Using Large Language Models

arXiv: https://arxiv.org/abs/2409.20364 · HTML: https://mlsanity.com/paper/2409.20364/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in Chain-of-thought reasoning for driving, one of 17 directions in LLM reasoning.
- Builds on 13 of its 23 references that are in the map.
- First posted on arXiv 30 Sep 2024.

## Placed in

- LLM reasoning / [Chain-of-thought reasoning for driving](https://mlsanity.com/lr/chain-of-thought-reasoning-for-driving/)
- End-to-end autonomous driving / [Large language models for driving](https://mlsanity.com/driving/large-language-models-for-driving/)

## Builds on (most cited first, 13 of 13)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Video-LLaMA: An Instruction-tuned Audio-Visual Language Model for Video Understanding](https://mlsanity.com/paper/2306.02858/) (2023)
- [Video-ChatGPT: Towards Detailed Video Understanding via Large Vision and Language Models](https://mlsanity.com/paper/2306.05424/) (2023)
- [VideoChat: Chat-Centric Video Understanding](https://mlsanity.com/paper/2305.06355/) (2023)
- [DrivingGaussian: Composite Gaussian Splatting for Surrounding Dynamic Autonomous Driving Scenes](https://mlsanity.com/paper/2312.07920/) (2023)
- [Vista: A Generalizable Driving World Model with High Fidelity and Versatile Controllability](https://mlsanity.com/paper/2405.17398/) (2024)
- [GenAD: Generalized Predictive Model for Autonomous Driving](https://mlsanity.com/paper/2403.09630/) (2024)
- [Mobile Edge Intelligence for Large Language Models: A Contemporary Survey](https://mlsanity.com/paper/2407.18921/) (2024)
- [DeepSpeed-FastGen: High-throughput Text Generation for LLMs via MII and DeepSpeed-Inference](https://mlsanity.com/paper/2401.08671/) (2024)
- [ADAPT: Action-aware Driving Caption Transformer](https://mlsanity.com/paper/2302.00673/) (2023)
- [Modeling Human Driving Behavior through Generative Adversarial Imitation Learning](https://mlsanity.com/paper/2006.06412/) (2020)
- [LimSim++: A Closed-Loop Platform for Deploying Multimodal LLMs in Autonomous Driving](https://mlsanity.com/paper/2402.01246/) (2024)
- [POLCA: Power Oversubscription in LLM Cloud Providers](https://mlsanity.com/paper/2308.12908/) (2023)

## Built on by (most cited first, 3 of 3)

- [Trajectory Mamba: Efficient Attention-Mamba Forecasting Model Based on Selective SSM](https://mlsanity.com/paper/2503.10898/) (2025)
- [Large Language Models and Their Applications in Roadway Safety and Mobility Enhancement: A Comprehensive Review](https://mlsanity.com/paper/2506.06301/) (2025)
- [BioLLMAgent: A Hybrid Framework with Enhanced Structural Interpretability for Simulating Human Decision-Making in Computational Psychiatry](https://mlsanity.com/paper/2603.05016/) (2026)

## Abstract (arXiv)

Deep learning architectures with powerful reasoning capabilities have driven significant advancements in autonomous driving technology. Large language models (LLMs) applied in this field can describe driving scenes and behaviors with a level of accuracy similar to human perception, particularly in visual tasks. Meanwhile, the rapid development of edge computing, with its advantage of proximity to data sources, has made edge devices increasingly important in autonomous driving. Edge devices process data locally, reducing transmission delays and bandwidth usage, and achieving faster response times. In this work, we propose a driving behavior narration and reasoning framework that applies LLMs to edge devices. The framework consists of multiple roadside units, with LLMs deployed on each unit. These roadside units collect road data and communicate via 5G NSR/NR networks. Our experiments show that LLMs deployed on edge devices can achieve satisfactory response speeds. Additionally, we propose a prompt strategy to enhance the narration and reasoning performance of the system. This strategy integrates multi-modal information, including environmental, agent, and motion data. Experiments conducted on the OpenDV-Youtube dataset demonstrate that our approach significantly improves performance across both tasks.