A driving Q-learning agent trained on language-structured scenarios develops threat-focused attention but its actions stay close to a constant baseline, a failure formalized as the recognition-control gap.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Language-Structured Relational Q-Learning for Threat-Aware Control in Safety-Critical Driving
A driving Q-learning agent trained on language-structured scenarios develops threat-focused attention but its actions stay close to a constant baseline, a failure formalized as the recognition-control gap.