This dataset contains both front-page and in-text citations from patents to scientific articles through 2020. If you use the data, please cite these two articles: 1. M. Marx & A. Fuegi, "Reliance on Science by Inventors: Hybrid Extraction of In-text Patent-to-Article Citations." forthcoming in Journal of Economics and Management Strategy. (http://doi.org/10.1111/jems.12455) 2. M. Marx, & A. Fuegi, "Reliance on Science: Worldwide Front-Page Patent Citations to Scientific Articles" (2020), Strategic Management Journal 41(9):1572-1594. (https://onlinelibrary.wiley.com/doi/full/10.1002/smj.3145) The datafile containing the citations is _pcs_mag_doi_pmid.tsv. DOIs and PMIDs provided where available. Each citation has the applicant/examiner flag, confidence score (1-10), and whether the reference was a) only on the front page, b) only in the body text, or c) in both. Each paper-patent citation also includes a preview release (think: alpha, not beta) of the temporal gap (in months) and three related measures of self-citation (i.e., was one or more of the inventors on the citing patent also an author on the cited paper). _data_description.pdf has full details. bodytextknowngood.tsv contains the known-good references for calculating recall. The remaining files redistribute much of the *final* edition of the Microsoft Academic Graph (12/20/2021). Please also cite Sinha, A, et al. 2015. Overview of Microsoft Academic Service (MAS) and Applications. In Proceedings of the 24th International Conference on World Wide Web (WWW ’15 Companion). ACM, New York, NY, USA, 243-246. Note that jif.zip, jcif.zip, and the OECD/wos-category crosswalks are derivatives of MAG and may not be updated through the end of 2021. These data are under an Open Data Commons Attribution license (ODC-By); use them for anything as long as you cite us! Source code for front-page matches is at https://github.com/mattmarx/reliance_on_science and for in-text is at https://github.com/mattmarx/intextcitations. Questions & feedback to firstname.lastname@example.org. This work is sponsored by the Alfred P. Sloan Foundation grant #G-2021-16822.