HanIFEval is a Korean translation of 429 items of google/IFEval (revision 966cd89). Every Korean prompt is translated together with the arguments of the rule-based checker that scores it, and the result is verified by code. as the v1 revision tag. Wherever the checker scores a constraint, the Korean prompt states the - the same string (keyword, forbidden word, end phrase, repeat text); - the same number; - the same relation ("N 이상", "N 미만", and "N개 이상" wherever the checker scores ≥ N). It does not mean that every natural reading of a prompt is scored. - Keyword rules match surface substrings. '청사진' contains '사진', and a spacing variant escapes a forbidden compound. - Some constraints are…
Independent publisher
Yonghye Kwon
developer0hye
Models in Library0
Datasets in Library1
Models on Hugging Face1
Followers2
Compare the published evidence
This profile connects the 1 model and dataset record shown below to their upstream cards and sources. The library counts describe SAVRN's coverage, rather than this publisher's entire output.
Published tasks: Text generation.
Reported licenses: apache-2.0. Open a record to review its specific license and conditions.
Compare model specifications, source dates, and hardware estimates on the individual records. Publisher identity verification is separate from a model's quality, deployment readiness, or suitability for your work.