Bench2ADVLM: A Closed-Loop Benchmark for Vision-language Models in Autonomous Driving

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Tianyuan, Jin, Ting, Wang, Lu, Liu, Jiangfan, Liang, Siyuan, Zhang, Mingchuan, Liu, Aishan, Liu, Xianglong
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908495139307520
author Zhang, Tianyuan
Jin, Ting
Wang, Lu
Liu, Jiangfan
Liang, Siyuan
Zhang, Mingchuan
Liu, Aishan
Liu, Xianglong
author_facet Zhang, Tianyuan
Jin, Ting
Wang, Lu
Liu, Jiangfan
Liang, Siyuan
Zhang, Mingchuan
Liu, Aishan
Liu, Xianglong
contents Vision-Language Models (VLMs) have recently emerged as a promising paradigm in autonomous driving (AD). However, current performance evaluation protocols for VLM-based AD systems (ADVLMs) are predominantly confined to open-loop settings with static inputs, neglecting the more realistic and informative closed-loop setting that captures interactive behavior, feedback resilience, and real-world safety. To address this, we introduce Bench2ADVLM, a unified hierarchical closed-loop evaluation framework for real-time, interactive assessment of ADVLMs across both simulation and physical platforms. Inspired by dual-process theories of cognition, we first adapt diverse ADVLMs to simulation environments via a dual-system adaptation architecture. In this design, heterogeneous high-level driving commands generated by target ADVLMs (fast system) are interpreted by a general-purpose VLM (slow system) into standardized mid-level control actions suitable for execution in simulation. To bridge the gap between simulation and reality, we design a physical control abstraction layer that translates these mid-level actions into low-level actuation signals, enabling, for the first time, closed-loop testing of ADVLMs on physical vehicles. To enable more comprehensive evaluation, Bench2ADVLM introduces a self-reflective scenario generation module that automatically explores model behavior and uncovers potential failure modes for safety-critical scenario generation. Overall, Bench2ADVLM establishes a hierarchical evaluation pipeline that seamlessly integrates high-level abstract reasoning, mid-level simulation actions, and low-level real-world execution. Experiments on diverse scenarios across multiple state-of-the-art ADVLMs and physical platforms validate the diagnostic strength of our framework, revealing that existing ADVLMs still exhibit limited performance under closed-loop conditions.
format Preprint
id arxiv_https___arxiv_org_abs_2508_02028
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Bench2ADVLM: A Closed-Loop Benchmark for Vision-language Models in Autonomous Driving
Zhang, Tianyuan
Jin, Ting
Wang, Lu
Liu, Jiangfan
Liang, Siyuan
Zhang, Mingchuan
Liu, Aishan
Liu, Xianglong
Computer Vision and Pattern Recognition
Vision-Language Models (VLMs) have recently emerged as a promising paradigm in autonomous driving (AD). However, current performance evaluation protocols for VLM-based AD systems (ADVLMs) are predominantly confined to open-loop settings with static inputs, neglecting the more realistic and informative closed-loop setting that captures interactive behavior, feedback resilience, and real-world safety. To address this, we introduce Bench2ADVLM, a unified hierarchical closed-loop evaluation framework for real-time, interactive assessment of ADVLMs across both simulation and physical platforms. Inspired by dual-process theories of cognition, we first adapt diverse ADVLMs to simulation environments via a dual-system adaptation architecture. In this design, heterogeneous high-level driving commands generated by target ADVLMs (fast system) are interpreted by a general-purpose VLM (slow system) into standardized mid-level control actions suitable for execution in simulation. To bridge the gap between simulation and reality, we design a physical control abstraction layer that translates these mid-level actions into low-level actuation signals, enabling, for the first time, closed-loop testing of ADVLMs on physical vehicles. To enable more comprehensive evaluation, Bench2ADVLM introduces a self-reflective scenario generation module that automatically explores model behavior and uncovers potential failure modes for safety-critical scenario generation. Overall, Bench2ADVLM establishes a hierarchical evaluation pipeline that seamlessly integrates high-level abstract reasoning, mid-level simulation actions, and low-level real-world execution. Experiments on diverse scenarios across multiple state-of-the-art ADVLMs and physical platforms validate the diagnostic strength of our framework, revealing that existing ADVLMs still exhibit limited performance under closed-loop conditions.
title Bench2ADVLM: A Closed-Loop Benchmark for Vision-language Models in Autonomous Driving
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2508.02028