KLUE

When this dataset fits

KLUE is a Korean NLP Dataset from Hugging Face (KLUE). It fits the korean nlp dataset task. Do not mix it with a benchmark download. License label CC BY-SA 4.0. Size text is “소용량”, copied from the host.

Open 3 catalog excerpts →

Task / category

Text · Text · NLI, KLUE

Language

Korean

Format · Size

Text · Small (25K sentences)

License

CC BY-SA 4.0(Catalog verified) · License document

Source & original link

Hugging Face (KLUE) · Original dataset link(Verified)

Verification status

Source: Verified · License: Verified · Last checked: 2026-09-03

KLUE: Korean NLP Dataset from Hugging Face (KLUE). License CC BY-SA 4.0. Open the original link to download.

KLUE (2021). Hugging Face (KLUE). license label: CC BY-SA 4.0 catalog commercial mark: allowed Original: https://huggingface.co/datasets/klue/klue. GearDel catalog: https://geardel.com/en/datasets/klue-nli. Last checked: 2026-09-03. Author names are not stored in the GearDel catalog and are not invented here.

Differences below use catalog fields only: license, commercial mark, provider, size text, and year.

• KorNLIKorean NLP Dataset · GitHub (kakaobrain)ViewProviders differ (Hugging Face (KLUE) vs GitHub (kakaobrain)). Size text differs (Small vs Large). Those strings are copied from the hosts. Catalog years differ (2021 vs 2020).
• KLUEKorean NLP Dataset · Hugging Face (KLUE)ViewThe catalog fields above match. Task and label differences still live on the original pages.
• KLUEKorean NER Dataset · Hugging Face (KLUE)ViewSize text differs (Small vs Medium). Those strings are copied from the hosts.
• KLUEKorean NLP Dataset · Hugging Face (KLUE)ViewThe catalog fields above match. Task and label differences still live on the original pages.

EXCERPTS

Catalog excerpts

3 stored samples. GearDel does not host the original files.

Open this row in Try samples →

어제 서울광장에는 수많은 시민들이 모여 평화 촛불 집회를 개최했습니다.

어제 서울광장에는 단 한 명의 시민도 나타나지 않았습니다.

Contradiction (모순)