论文名称:Efficient Sequential Policy Optimization via Off-Policy Correction in Multi-Agent Reinforcement Learning
搜索显示三位作者均隶属于中国科学院大学。
审稿人意见:该论文因与 [1] 有大量文字重叠而被否决。这一决定得到了项目委员会多位成员的确认。本论文中的第 5 节是 [1] 中第 2 节的轻微改写,第 2.1 节与 [1] 中的第 3.1 节几乎相同,第 4 节与 [1] 中的第 5 节几乎相同,摘要是 [1] 中摘要的改写。 因此,重叠的程度被视为抄袭,在抄袭整个文本/短语等时没有给出适当的引文。
The paper is desk rejected for significant textual overlap with [1]. This decision was confirmed by multiple members of the program committee. Section 5 in this submission is a slight rewording of Section 2 in [1], Section 2.1 in this submission is nearly identical to section 3.1 in [1], Section 4 is nearly identical to Section 5 in [1], and the abstract is a rewording of the abstract in [1]. As such, the degree of overlapping is considered as plagiarism where the proper citation is not given when the whole texts/phrases etc are copied. [1] Wang, X., Tian, Z., Wan, Z., Wen, Y., Wang, J., & Zhang, W. (2023). Order Matters: Agent-by-agent Policy Optimization. ICLR 2023.
原文地址:https://www.zhihu.com/question/1235573602 |