MarathiEmoExplain

MarathiEmoExplain is a 10,762-sentence Marathi dataset annotated with sentiment, fine-grained emotion, and human-written justifications. It enables both classification and reasoning generation benchmarks, supporting interpretable, culturally grounded NLP research in a low-resource setting.

Citation

If you use the MarathiEmoExplain dataset in your research, please cite the following paper:

@inproceedings{kumar-etal-2025-marathiemoexplain,
    title = "{M}arathi{E}mo{E}xplain: A Dataset for Sentiment, Emotion, and Explanation in Low-Resource {M}arathi",
    author = "Kumar, Anuj  and
      Sayed, Mohammed Faisal  and
      Ahlawat, Satyadev  and
      Prasad, Yamuna",
    booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2025",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2025.findings-emnlp.712/",
    doi = "10.18653/v1/2025.findings-emnlp.712",
    pages = "13234--13243",
    ISBN = "979-8-89176-335-7",
}