Post by Paul G. Allen School of Computer Science & Engineering

Web archives like the one that preserves federal websites from past presidential administrations (known, fittingly, as the End of Term Web Archive), contain millions of PDFs preserving important information about our nation’s history and governance. It turns out that preservation is the easy part; being able to search and retrieve data contained in the archive is not. But that’s changing thanks to a team of University of Washington researchers that includes #UWAllen alumni Benjamin Lee (Ph.D., ‘23) and Kyle Deeds (Ph.D., ‘25). They recently introduced a new open-source #AI tool, GovScape, that enables historians, journalists and other interested members of the public to discover relevant documents within these vast archives using new multimodal search capabilities. These include semantic search, which allows you to search for a topic like “endangered species protection” without knowing the exact terms the relevant documents might include, and visual search, which can be used to find charts and images or exhibiting qualities such as redaction. “I’m really excited about the prospects for better access to government information with projects like GovScape,” said Lee, who is now a faculty member in the University of Washington Information School. “Being able to actually find relevant information is vital to the health of democracy and to the functioning of society.” In addition to Lee and Deeds, Allen School undergraduate researchers Claire Gong, Shreya Shaji, Alison Yan, Albert Du and Anjali Gopal contributed to the project, which the team presented at the Annual Meeting of the Association for Computational Linguistics in San Diego earlier this month. #ArtificialIntelligence #AIforGood #OpenGovernment #UndergraduateResearch #ACL2026 #UWdiscovers Read the UW News story: https://lnkd.in/erD3tnqX