China

REARL, a closed-loop simulation enhancement framework that integrates real traffic data with large language model (LLM) and modulates the simulated vehicle with reference to that real action, is proposed.

Personalized evaluation rubrics reveal that user satisfaction can diverge significantly from generic quality assessments in role-playing agents.

MLLMs can now reason about streaming video with significantly improved accuracy and reduced output length thanks to a novel memory-anchored framework that overlaps watching and thinking.