VLA driving models exhibit only 42.5% reasoning fidelity, miss pedestrians frequently, show 97.7% trajectory fragility to mild perturbations, and display low reasoning-action consistency in nearly half of cases.
Training language models to follow in- structions with human feedback
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Is VLA Reasoning Faithful? Probing Safety of Chain-of-Causation
VLA driving models exhibit only 42.5% reasoning fidelity, miss pedestrians frequently, show 97.7% trajectory fragility to mild perturbations, and display low reasoning-action consistency in nearly half of cases.