AI Paper Atlas

Research map / Vision-language and multimodal / Robot action and navigation

DrivingGPT: Unifying Driving World Modeling and Planning with Multi-modal Autoregressive Transformers

Yuntao Chen, Yuqi Wang, Zhaoxiang Zhang

ICCV 2025 · Vision-language and multimodal › Robot action and navigation · approach: Embodied multimodal LLM agents · 68 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace diffusion SVD fine-tuning with autoregressive DrivingGPT for video generationvideo world model · navtest video generationFVD: 227.54 -> 142.61Improved
replace diffusion SVD fine-tuning with autoregressive DrivingGPT for video generationvideo world model · navtest video generationFID: 24.03 -> 12.78Improved
use DrivingGPT instead of SVD for 64-frame long video generationvideo world model · navtest long video generation (64 frames)FVD: 1079.28 -> 506.95Improved
fine-tune SVD on navtrain instead of using it zero-shotSVD · navtest video generationFVD: 483.76 -> 227.54Improved
keep predicted longitudinal x action instead of copying last history action (ablation)DrivingGPT planner · NAVSIM navmini planningPDMS: 53.5 -> 82.4Improved
keep predicted lateral y action instead of copying last history action (ablation)DrivingGPT planner · NAVSIM navmini planningPDMS: 79.4 -> 82.4Improved
keep predicted yaw action instead of copying last history action (ablation)DrivingGPT planner · NAVSIM navmini planningPDMS: 73.1 -> 82.4Improved
train on curated NAVSIM data (104k sequences) rather than 651k nuPlan uniform sequences (ablation)DrivingGPT planner · NAVSIM navmini planningPDMS: 74.6 -> 82.4Improved
add 1D rotary position embedding to action tokens as well as image tokens (ablation)DrivingGPT planner · NAVSIM navmini planningPDMS: 65.3 -> 82.4Improved
use Llama-Gen tokenizer instead of OpenMAGVIT2 (ablation)DrivingGPT visual tokenizer · navtest tokenizer reconstructionrFID: 6.70 -> 5.67Improved
use Llama-Gen tokenizer instead of Cosmos Tokenizer (ablation)DrivingGPT visual tokenizer · navtest tokenizer reconstructionPSNR: 25.05 -> 23.09Mixed
DrivingGPT without auxiliary supervision versus LAWDrivingGPT · NAVSIM navmini planningPDMS: 82.7 -> 82.4No clear effect

Later papers that compared with this method

1 later accepted paper reports DrivingGPT as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy