AI Paper Atlas

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

AutoMoT: A Unified Vision-Language-Action Model with Asynchronous Mixture -of-Transformers for End-to-End Autonomous Driving

Wenhui Huang, Songyan Zhang, Qihang Huang, Zhidong Wang, Zhiqi Mao, Collister Chua, Zhan Chen, Long Chen, Chen Lv

ICML 2026 · Vision-language and multimodal › Robot action and navigation · approach: Vision-language-action foundation models · 0 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
add diffusion-based action refiner (AR) on top of AutoMoT (ablation)AutoMoT · Bench2Drive closed-loopDriving Score: 87.34 -> 89.42Improved
add diffusion-based action refiner (AR) on top of AutoMoT (ablation)AutoMoT · Bench2Drive closed-loopSuccess Rate (%): 70.00 -> 74.09Improved
fine-tune VLM backbone on LingoQA and OmniDrive counterfactual data (ablation)AutoMoT understanding expert · LingoQALingo-Judge: 67.00 -> 67.20No clear effect
fine-tune VLM backbone on AD data (ablation)AutoMoT understanding expert · OmniDrive counterfactual planningGPT score: 18.20 -> 67.80Improved
fine-tune VLM backbone on AD data (ablation)AutoMoT understanding expert · TallyQAtoken accuracy: 81.40 -> 52.40Hurt
fine-tune VLM backbone on AD data (ablation)AutoMoT understanding expert · InfographicVQAGPT score: 89.30 -> 50.20Hurt
fine-tune VLM backbone on AD data (ablation)AutoMoT understanding expert · VizWizGPT score: 75.60 -> 50.20Hurt
fine-tune VLM backbone on AD data (ablation)AutoMoT understanding expert · FigureQAtoken accuracy: 97.60 -> 91.20Hurt
fine-tune VLM backbone on AD data (ablation)AutoMoT understanding expert · ScienceQAtoken accuracy: 88.60 -> 87.80No clear effect
use asynchronous inference with persistent KV cache instead of synchronized UE/AE (ablation)AutoMoT · nuScenes trajectory planningAvg L2 (m): 0.322 -> 0.324No clear effect
use asynchronous inference with persistent KV cache instead of synchronized UE/AE (ablation)AutoMoT · Inference latency without diffusion headtotal latency (ms): 117.3 -> 37.0Improved
use asynchronous inference with persistent KV cache instead of synchronized UE/AE (ablation)AutoMoT with diffusion head · Inference latency with diffusion headtotal latency (ms): 143.3 -> 63.0Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 4 of 5 on most shared results.

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