TY - RPRT TI - RageVLA: Augmentation-Trained Perception and Action for State-of-the-Art Robustness on LIBERO AU - Amrit Lal Singh AU - Smarak Kanjilal PY - 2026 UR - https://arxiv.science/abs/2611.23517 ID - 2611.23517 N1 - Source identifier: 2611.23517 ER -