LATIDIA · Robótica

LIBERO-Pro2.0: un punto de referencia para evaluar la solidez de los modelos de cimentación incorporados

arXiv: 2610.24587v1Tipo de anuncio: nuevo Resumen: LIBERO se ha convertido en un punto de referencia ampliamente adoptado para evaluar las políticas de manipulación robótica. Sin embargo, un buen rendimiento en LIBERO no indica necesariamente un sólido

La noticia

arXiv:2610.24587v1 Announce Type: new Abstract: LIBERO has emerged as a widely adopted benchmark for evaluating robotic manipulation policies. However, strong performance on LIBERO does not necessarily indicate robust embodied capabilities, as models may exploit benchmark-specific regularities rather than generalize to diverse operating conditions. In this work, we introduce LIBERO-Pro2.0, a comprehensive benchmark for systematically evaluating the robustness of embodied foundation models under distribution shifts. LIBERO-Pro2.0 encompasses 22 static perturbations and 20 runtime interventions across six dimensions: Condition, Environment, Observation, Execution, Robot State Configuration, and Language, yielding 1,648 validated shifted task instances. To support robustness-oriented learning, we further collect 39,390 training trajectories through model-generated and human demonstrations. All shifted task instances