MTEB logo MTEB
classification text

IndicLangClassification

Cite this task

@inproceedings{madhani-etal-2023-bhasa,
  address = {Toronto, Canada},
  author = {Madhani, Yash  and
Khapra, Mitesh M.  and
Kunchukuttan, Anoop},
  booktitle = {Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)},
  doi = {10.18653/v1/2023.acl-short.71},
  editor = {Rogers, Anna  and
Boyd-Graber, Jordan  and
Okazaki, Naoaki},
  month = jul,
  pages = {816--826},
  publisher = {Association for Computational Linguistics},
  title = {Bhasa-Abhijnaanam: Native-script and romanized Language Identification for 22 {I}ndic languages},
  url = {https://aclanthology.org/2023.acl-short.71},
  year = {2023},
}

A language identification test set for native-script as well as Romanized text which spans 22 Indic languages.

Domains: WebNon-fictionWritten
Languages: AssameseBengaliBodo (India)Dogri (macrolanguage)Goan KonkaniGujaratiHindiKannadaKashmiriMaithiliMalayalamMarathiManipuriNepali (individual language)OdiaPanjabiSanskritSantaliSindhiTamilTeluguUrdu
Task type
Classification
Main metric
accuracy
License
cc0-1.0
Dates
2022-08-01 → 2023-01-01
Annotations
expert-annotated
Sample creation
created
Models scored …
Dataset preview mteb/IndicLangClassification Open on HuggingFace →

Model scores

Loading…