SYNOPSIS: Using a machine-assisted human approach, we provide a comprehensive review of 1,041 IFRS articles published in top-tier journals (ABS 4*, 4, and 3) between 1984 and 2025. We combine latent Dirichlet allocation (LDA), a probabilistic topic-modeling technique, with human validation and interpretation to map the thematic structure and development of IFRS research. We identify ten major IFRS research topics and examine the studies that have shaped these research streams and the methodological approaches underlying their development. Citation patterns indicate that research on the topics Economic and financial consequences of IFRS adoption and Convergence and harmonization-related issues in IFRS has attracted particularly strong scholarly attention. Moreover, we document publication trends over the years across several dimensions and identify influential authors and journals in each area. Finally, we assess the usefulness of generative large language models for mapping the IFRS literature, emphasizing both their analytical promise and their methodological limitations. Data Availability: Data used in this paper are from publicly available sources. JEL Classifications: M41; C88; G38.