MuLTiGENBiaS

Automatic Identification of Gender Bias in Hindi,Bengali,Meitei Codemixed Texts

This is a XLM-Align-Base model trained on CoMMA dataset of 12k samples

• This is an extension work from our previous paper: Hypers at ComMA@ICON: Modelling Aggressiveness, Gender Bias and Communal Bias Identification.

Example Usage

import torch
import numpy as np
from transformers import AutoTokenizer, AutoModelForSequenceClassification, pipeline
from transformers import set_seed

set_seed(425)

text = "some gender biased text"
pipe = pipeline("text-classification", model="seanbenhur/MuLTiGENBiaS")


def predict_pipe(text):
    prediction = pipe(text, return_all_scores=True)[0]
    return prediction


if __name__ == "__main__":
  target = predict_pipe(text)
  print(target)
  

Some concerns

• Note: The model is trained on relatively lower samples (i.e 12k) but with mix of four languages Hindi, Bengali, Meitei, and English. It contains both native on codemixed scripts, So the model might perform poorly on many text samples and might not generalize well.

Bibtex

 @article{Benhur2021HypersAC,
  title={Hypers at ComMA@ICON: Modelling Aggressiveness, Gender Bias and Communal Bias Identification},
  author={Sean Benhur and Roshan Nayak and Kanchana Sivanraju and Adeep Hande and Subalalitha Chinnaudayar Navaneethakrishnan and Ruba Priyadharshini and Bharathi Raja Chakravarthi6},
  journal={ArXiv},
  year={2021},
  volume={abs/2112.15417}
}