Mobile learning strategies have been employed for social learning activities, including library- and museum-supported learning. Previous studies have reviewed the literature from the technological aspect. However, a retrospective study from the perspective of bibliometric and network structure has not yet been provided. The aim of this study was therefore to systematically review journal papers on library-supported mobile learning (LibML). A coding framework including library types, mobile learning strategies, and research issues was adopted based on the literature and was used to screen and categorize the research papers. A co-citation network analysis was then adopted to analyze and visualize the structural relationships among the papers. A total of 53 eligible articles with 1370 citations in follow-up studies were collected from the Scopus database. The results showed that two main research streams of LibML were identified from the overall network structure, including library- and museum-supported mobile learning. In terms of the mobile learning strategy, library-supported research mainly focused on self-directed learning, whereas museum-supported research emphasized inquiry-based learning. In terms of research issues, most library-supported research focused on patrons' affective engagement, whereas museum-supported research emphasized learning performance. This study provides a citation-based approach to reveal the research trends and mainstream LibML research. The main contribution of combining co-citation and social network analysis is to provide a visualized network diagram of LibML research. Limitations of the methodological approach are noted. Discussion and future directions from the follow-up study are provided.