SOURCE-LINKED INTELLIGENCE

Imagine-RL: Residual-Confidence-Guided Cross-Attention for World-Model-Augmented VLA Reinforcement Learning

arXiv · AI, language, vision and robotics · article · Sep 21, 2026 · UTC

Reliable action evaluation in contact-rich manipulation requires looking beyond the current observation to future visual and contact consequences. Existing noise-space reinforcement learning efficiently steers a frozen Vision-Language-Action (VLA) policy, but its critics largely ignore these consequences. We present Imagine-RL, which augments noise-space VLA post-training with action-conditioned visual-torque imagination. For each candidate action chunk, a frozen visual-torque latent world model (VTLWM) autoregressively predicts compact future representations without pixel reconstruction. A cu

Read original source ↗ Open in workspace

recordType

paper

region

Global