Package: evaldrift
Type: Package
Title: Paraphrase-Drift Scoring for LLM Eval Sets
Version: 0.1.0
Authors@R: person("James", "Verlander", email = "scrapertweeter3@gmail.com", role = c("aut", "cre"))
Maintainer: James Verlander <scrapertweeter3@gmail.com>
Description: Paraphrase-heavy eval sets drift over time: answers get
        rewritten, questions get reworded, and nobody re-runs the
        comparison that justified keeping both versions. eval_drift()
        scores two texts with three plain similarity measures
        (word-shingle Jaccard, character-trigram Jaccard, normalized
        Levenshtein similarity) and one blended drift score. No model
        calls, no network access, pure base R. Companion to the
        eval-methodology writing on PastAGI <https://pastagi.com>.
License: MIT + file LICENSE
Encoding: UTF-8
URL: https://pastagi.com
BugReports: https://github.com/scrapertweeter3-prog/evaldrift/issues
RoxygenNote: 7.3.1
Repository: https://scrapertweeter3-prog.r-universe.dev
Date/Publication: 2026-10-21 23:19:50 UTC
RemoteUrl: https://github.com/scrapertweeter3-prog/evaldrift-r
RemoteRef: HEAD
RemoteSha: 67b0f0530be28a47f11a57eba0680fd8f3549a7b
NeedsCompilation: no
Packaged: 2026-10-22 00:08:34 UTC; root
Author: James Verlander [aut, cre]