https://penn-pal-lab.github.io/LIV/ LIV LIV as Representation for Language-Conditioned BC We use LIV's frozen multi-modal representation as backbone for LCBC and achieve impressive performance (46% success rate, absolute ~30% better than the second best baseline) on a challenging real-world muli penn-pal-lab.github.io 설명: https://twitter.com/JasonMa2020/status/1663618652778942464 트위터에서 즐기는 Jason..