2026 Fall DLC: Keynote Address: GovScape: Toward Multimodal Search & Discovery for Government PDFs

Widespread efforts by libraries, archives, and museums have drastically improved digital access to collections. Yet, scholars and the public alike face a persistent challenge: how to explore and analyze these collections, which contain millions or even billions of items and often suffer from incomplete metadata. My interdisciplinary research addresses this question by bringing together approaches from AI, library & information science, and the digital humanities in order to build and examine large-scale search systems for digital collections. In this talk, I will introduce GovScape, a multimodal search system for 10+ million Government PDFs publicly available at: https://govscape.net. I will conclude with a horizon of opportunities and challenges that lie ahead with rendering Government documents discoverable in the age of AI.