LIBERO-PRO: Towards Robust and Fair Evaluation of Vision-Language-Action Models Beyond Memorization

Zhou, Xueyang, Xu, Yangming, Tie, Guiyao · arXiv · 2025