ICR-Drive reveals substantial performance drops in end-to-end language-driven driving models when instructions are paraphrased, made ambiguous, noised, or misleading.
Are vlms ready for autonomous driving? an empirical study from the reliability, data and metric perspectives
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
ICR-Drive: Instruction Counterfactual Robustness for End-to-End Language-Driven Autonomous Driving
ICR-Drive reveals substantial performance drops in end-to-end language-driven driving models when instructions are paraphrased, made ambiguous, noised, or misleading.