Jul 2026
Self-Evolving Learning for Embodied AI with Criticality Model
This work proposes a self-evolving method that reduces failure rates by 51--67% relative to trained baselines and by 8-25% relative to state-of-the-art vision-language-action models after replacing redundant nominal scenarios with diverse failure-prone ones.
Linxuan He, Yuying Tian, Ling-Xiang Fan et al.
· arXiv.org · 0 citations