Metrics for Assessing the Quality of Generated Text R package sacRebleu version 0.2.0 Implementation of the BLEU-Score in 'C++' to evaluate the quality of generated text. The BLEU-Score, introduced by Papineni et al. (2002) <doi:10.3115/1073083.1073135>, is a metric for evaluating the quality of generated text. It is based on the n-gram overlap between the generated text and reference texts. Additionally, the package provides some smoothing methods as described in Chen and Cherry (2014) <doi:10.3115/v1/W14-3346>. Software Depends: R (>= 4.2.0) Imports: checkmate, Rcpp (>= 1.0.12) LinkingTo: Rcpp Suggests: knitr, rmarkdown, testthat (>= 3.0.0), vctrs, withr Philipp Koch <PhillKoch@protonmail.com> Comprehensive R Archive Network (CRAN) Philipp Koch [aut, cre] GPL (>= 2) 2025-01-22 application/tgz https://CRAN.R-project.org/package=sacRebleu doi:10.32614/CRAN.package.sacRebleu en-US