Showing cs.ROShow all
2 papers · 1 filter
cs.RO2026
Imagine-RL: Residual-Confidence-Guided Cross-Attention for World-Model-Augmented VLA Reinforcement Learning
Kejia Hu, Wentong Zhai, Bo Zhao +1
Reliable action evaluation in contact-rich manipulation requires looking beyond the current observation to future visual and contact consequences. Existing noise-space reinforcemen…
cs.RO2026
PAVE: Predictive Alignment and Value-Guided Evolution for World-Action Policies
Botong Zhao, Fang Yu, Tim Yu +3
Direct vision-language-action policies generate continuous robot actions efficiently, but standard behavior cloning leaves two complementary gaps: their representations are not exp…