Bench2ADVLM: A Closed-Loop Benchmark for Vision-language Models in Autonomous Driving
This addresses the need for more robust safety assessment in autonomous driving systems, though it is incremental as it focuses on evaluation rather than model improvement.
The paper tackles the lack of realistic closed-loop evaluation for vision-language models in autonomous driving by introducing Bench2ADVLM, a hierarchical framework that enables testing in both simulation and on physical vehicles, revealing that current models perform poorly under these conditions.
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.