MetaGraph: A Large-Scale Meta-Analysis of GenAI in Financial NLP (2022–2025) This dataset accompanies the paper “MetaGraph: A Large-Scale Meta-Analysis of GenAI in Financial NLP (2022–2025)”. It introduces MetaGraph, a machine-readable knowledge graph automatically constructed from 681 scientific papers in the field of GenAI-driven Financial NLP. MetaGraph enables large-scale, structured meta-analyses of research trends, conceptual relationships, and method co-usage in Financial NLP, with a focus on the rapid transformation of the field since late 2022. Content Overview MetaGraph represents the FinNLP research landscape as a structured graph where: Nodes correspond to papers, authors, institutions, models, datasets, tasks, techniques, motivations, and limitations. Edges encode typed relationships among these entities, including paper–entity links, task–dataset–model relationships, authorship, institutional affiliation, and inferred co-occurrence patterns. The graph facilitates: Discovery of underexplored intersections in Financial NLP research Temporal analyses of GenAI adoption and field maturation Large-scale analysis of shifts in tasks, datasets, risks, and system design Visual exploration of conceptual and methodological trends Included Files finnlp_ontology.json: Defines an extensible ontology of Financial NLP concepts, entity types, attributes, and relation types. finnlp_graph.graphml: The full MetaGraph in GraphML format, compatible with tools such as Gephi, NetworkX, and GraphViz. finnlp_graph.json: A simplified JSON version for human-readable access, including structured metadata for nodes and edges. Use Cases This dataset is intended for researchers in NLP, finance, scientometrics, and scientific knowledge extraction. It can be used to: Conduct bibliometric and scientometric studies Analyze the methodological evolution of GenAI in Financial NLP Study temporal shifts in tasks, datasets, models, techniques, motivations, and limitations Train or evaluate systems for scientific knowledge extraction, graph construction, or research trend monitoring Citation If you use this resource, please cite the accompanying paper: Pedinotti, P., Baumann, P., Jessurun, N., Barrett, L., & Santus, E. (2026). MetaGraph: A Large-Scale Meta-Analysis of GenAI in Financial NLP (2022–2025). Proceedings of the Fifth Workshop on Generation, Evaluation and Metrics (GEM). Note: 58 of the 681 papers analyzed in the study are not included in the published dataset. These papers were posted to arXiv under a CC BY-NC-ND 4.0 license, which prohibits redistribution of derivative or adapted forms of the original material. As such, they are excluded from the released MetaGraph knowledge graph.