AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

Are VLMs Ready for Autonomous Driving? An Empirical Study from the Reliability, Data and Metric Perspectives

Shaoyuan Xie, Lingdong Kong, Yuhao Dong, Chonghao Sima, Wenwei Zhang, Qi Alfred Chen, Ziwei Liu, Liang Pan

ICCV 2025 · Vision-language and multimodal › Multimodal LLMs · approach: VLM capability analysis · 107 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
remove the image so the prompt is text-only (ablation)GPT-4o · DriveBench behaviorGPT score: 45.40 -> 50.03No clear effect
remove the image so the prompt is text-only (ablation)GPT-4o · DriveBench perceptionGPT score: 35.37 -> 36.48No clear effect
add the image back to the prompt (vs text-only) (ablation)Qwen2-VL-72B · DriveBench predictionGPT score: 5.57 -> 49.35Improved
add the image back to the prompt (vs fully black) (ablation)Qwen2-VL-72B · DriveBench perception MCQaccuracy: 23.5 -> 60.0Improved
add the image back to the prompt (vs fully black) (ablation)GPT-4o · DriveBench perception MCQaccuracy: 59.5 -> 59.0No clear effect
add the image back to the prompt (vs fully black) (ablation)Qwen2-VL-72B · DriveBench behavior MCQaccuracy: 36.5 -> 23.0Hurt
add the image back to the prompt (vs fully black) (ablation)Phi-3 · DriveBench perception MCQaccuracy: 17.5 -> 54.5Improved
apply corruptions to images (avg of 15)Human annotators · DriveBench behavior MCQGPT score: 69.51 -> 54.09Hurt
apply corruptions to images (avg of 15)GPT-4o · DriveBench perceptionGPT score: 35.37 -> 35.25No clear effect
name the corruption type explicitly in the question (ablation)LLaVA-NeXT · DriveBench MCQ (Snow)accuracy change: -31.95Hurt
name the corruption type explicitly in the question (ablation)GPT-4o · DriveBench MCQ (Dark)accuracy change: -12.98Hurt
name the corruption type explicitly in the question (ablation)LLaVA-1.5-7B · DriveBench MCQ (Bright)accuracy change: 0.26No clear effect

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