PASTA: Pretrained Action-State Transformer AgentsRaphael Boige,Yannis Flet-Berliac,Arthur Flajolet,Guillaume Richard,Thomas PierrotICLR 2024(2024)引用 1|浏览114关键词self-supervised pre-training,transformer models,reinforcement learningAI 理解论文溯源树样例生成溯源树,研究论文发展脉络Chat Paper正在生成论文摘要