Naoya Iwata
Core Member
Associate Professor, Center for Digital Humanities and Social Sciences, Nagoya University
Research Associate, National Institute of Informatics
M.A. from Kyoto University; Ph.D. in Ancient Philosophy from the University of Cambridge (2015). After a JSPS Overseas Research Fellowship at Cambridge and a British Academy Postdoctoral Fellowship at the University of Oxford, he joined the Graduate School of Humanities at Nagoya University as Associate Professor in 2023 and the Center for Digital Humanities and Social Sciences in 2024, with a concurrent appointment at the National Institute of Informatics since 2025. He works on ancient philosophy, especially Plato and Aristotle, with articles in the Journal of the History of Philosophy and the British Journal for the History of Philosophy, among others. Since 2023 he has been building Humanitext, a research infrastructure connecting classical texts and AI. He is principal investigator of the Schmidt Sciences HAVI project (2026–2029) and of the JST BOOST Next-Generation AI Researcher Development Program project (2025–2030). He is Program Committee Chair of JADH2026 and Organizing Committee Chair of JADH2027, the annual conferences of the Japanese Association for Digital Humanities, and received the IPSJ Yamashita SIG Research Award (2025).
As the leader of the Humanitext project, he is responsible for its overall design, development and operation, and is involved in the following research:
-
A knowledge graph of relationships between classical texts (Schmidt Sciences HAVI)
- Principal investigator of “Humanitext: A Human-AI Co-Creation Platform for Verified Classical Scholarship,” which began in October 2026. AI proposes relationships between classical texts — quotation, allusion, paraphrase and criticism — and experts verify each one before it is stored in a knowledge graph. He designs the research on intertextuality, leads RAG/GraphRAG development and serves as a final expert verifier. See the announcement.
-
Developing and running the applications
- He develops and runs Humanitext Antiqua (launched 2024), which answers questions about Greek and Latin classics by citing the source passages, together with Aozora for Japanese literature in the Aozora Bunko collection, an OCR system built on multimodal large language models, and Reader, for reading classical texts sentence by sentence with translations.
-
Making classical texts usable by AI
- His work includes a data design that splits TEI/XML texts according to the citation conventions of classical scholarship (CTS), so that retrieval keeps every passage traceable to its source (DH2026); “context-oriented translation,” which improves semantic search by rendering the original text into context-aware translations (IPSJ Yamashita SIG Research Award, 2025); and methods for evaluating retrieval accuracy in humanities RAG.
-
Extending to other fields
- He applies the methods developed in Humanitext to other kinds of sources, including Digital Digesta, which digitizes Justinian’s Digest of Roman law with AI-based OCR, and Humanitext Planetarium, which builds constellation narrations from classical sources.
-
Bringing the classics to a wider public
- Through ANIMA ANTIQUA, which turns scenes from classical texts into short films based on the original sources (Nenone Program, Tokai National Higher Education and Research System; DH2026 poster), and through articles and explainer videos on Greek mythology, he creates ways into the original texts for a general audience.