GovScape: A Public Multimodal Search System for 70 Million Pages of Government PDFs
Published in ACL 2026 (System Demonstrations), 2025
GovScape is a public search system supporting metadata, exact text, semantic text, and visual search across roughly 10 million federal government PDFs (70 million pages) drawn from the 2020 End of Term web crawl. Read more about the launch here.
Download paper here Try GovScape here
Recommended citation: Huang, Y.-H., Gong, C., Shaji, S., Yan, A. R., Harka, L., Du, A., Gopal, A. S., Klein, S. J., Shen, S. Z., Phillips, M. E., Owens, T., Deeds, K., & Lee, B. C. G. (2026). GovScape: A Public Multimodal Search System for 70 Million Pages of Government PDFs. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 3: System Demonstrations), 231-241. https://doi.org/10.18653/v1/2026.acl-demo.23
Recommended citation: Huang, Y.-H., Gong, C., Shaji, S., Yan, A. R., Harka, L., Du, A., Gopal, A. S., Klein, S. J., Shen, S. Z., Phillips, M. E., Owens, T., Deeds, K., & Lee, B. C. G. (2026). GovScape: A Public Multimodal Search System for 70 Million Pages of Government PDFs. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 3: System Demonstrations), 231-241. https://doi.org/10.18653/v1/2026.acl-demo.23
Download Paper