ICR-Drive: Instruction Counterfactual Robustness for End-to-End Language-Driven Autonomous Driving

TL;DR

ICR-Drive is a diagnostic framework that tests the robustness of language-conditioned autonomous driving models against instruction variations, revealing significant performance drops under minor perturbations.

Contribution

The paper introduces ICR-Drive, a novel method for systematically evaluating instruction robustness in autonomous driving models using controlled perturbations.

Findings

Minor instruction changes cause substantial performance drops.

Different perturbation types lead to distinct failure modes.

Robustness gaps highlight safety concerns for deployment.

Abstract

Recent progress in vision-language-action (VLA) models has enabled language-conditioned driving agents to execute natural-language navigation commands in closed-loop simulation, yet standard evaluations largely assume instructions are precise and well-formed. In deployment, instructions vary in phrasing and specificity, may omit critical qualifiers, and can occasionally include misleading, authority-framed text, leaving instruction-level robustness under-measured. We introduce ICR-Drive, a diagnostic framework for instruction counterfactual robustness in end-to-end language-conditioned autonomous driving. ICR-Drive generates controlled instruction variants spanning four perturbation families: Paraphrase, Ambiguity, Noise, and Misleading, where Misleading variants conflict with the navigation goal and attempt to override intent. We replay identical CARLA routes under matched simulator configurations and seeds to isolate performance changes attributable to instruction language. Robustness is quantified using standard CARLA Leaderboard metrics and per-family performance degradation relative to the baseline instruction. Experiments on LMDrive and BEVDriver show that minor instruction changes can induce substantial performance drops and distinct failure modes, revealing a reliability gap for deploying embodied foundation models in safety-critical driving.