Abstract: In recent years, graph representation learning methods have attracted significant attention for their ability to effectively handle graph-structured data. Chinese herbal medicine (CHM), with its multi-component, multi-target, and multi-pathway characteristics, demonstrates significant advantages in the treatment of complex diseases, particularly as different combinations of Chinese herbs can produce unique synergistic effects. Graph representation learning provides a new perspective for the in-depth study of CHM combinations. This paper first outlines the relevant methods of graph representation learning, explores the current application status of these methods in CHM combinations, and discusses the challenges and corresponding solutions. By reviewing the research dynamics and cutting-edge trends in this field, this paper aims to provide valuable references and insights for future in-depth research.