English Humor Detection Model

This model is part of the Humor Research collection of models for English humor detection, humor classification, and joke detection. It can be used for binary text classification tasks such as identifying whether an English text is humorous or non-humorous.

Model name

The model name indicates the dataset from the paper on which the model was trained. The numbers in the model name correspond to the random seed used for model initialization.

Recommended model for humor detection

If you need a single recommended model for English humor vs. non-humor classification, please refer to the best model from the project:

Humor-Research/humor-detection-comb-23

Paper

This model was released as part of the study:

You Told Me That Joke Twice: A Systematic Investigation of Transferability and Robustness of Humor Detection Models

GitHub repository

Code, data processing tools, and additional project information are available here:

Humor-Research/Humor-detection

Citation

If you use this model, please cite the following paper:

@inproceedings{baranov-etal-2023-told,
    title = "You Told Me That Joke Twice: A Systematic Investigation of Transferability and Robustness of Humor Detection Models",
    author = "Baranov, Alexander  and
      Kniazhevsky, Vladimir  and
      Braslavski, Pavel",
    editor = "Bouamor, Houda  and
      Pino, Juan  and
      Bali, Kalika",
    booktitle = "Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing",
    month = dec,
    year = "2023",
    address = "Singapore",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2023.emnlp-main.845",
    doi = "10.18653/v1/2023.emnlp-main.845",
    pages = "13701--13715",
}