Reducing Gender Bias in Word-Level Language Models with a Gender-Equalizing Loss Function

@inproceedings{Qian2019ReducingGB,
  title={Reducing Gender Bias in Word-Level Language Models with a Gender-Equalizing Loss Function},
  author={Yusu Qian and Urwa Muaz and Ben Zhang and Jae Won Hyun},
  booktitle={ACL},
  year={2019}
}
Gender bias exists in natural language datasets which neural language models tend to learn, resulting in biased text generation. In this research, we propose a debiasing approach based on the loss function modification. We introduce a new term to the loss function which attempts to equalize the probabilities of male and female words in the output. Using an array of bias evaluation metrics, we provide empirical evidence that our approach successfully mitigates gender bias in language models… CONTINUE READING

Citations

Publications citing this paper.

Reducing Sentiment Bias in Language Models via Counterfactual Evaluation

VIEW 3 EXCERPTS
CITES METHODS & BACKGROUND

References

Publications referenced by this paper.
SHOWING 1-10 OF 10 REFERENCES

Identifying and Reducing Gender Bias in Word-Level Language Models

VIEW 6 EXCERPTS
HIGHLY INFLUENTIAL

Learning Gender-Neutral Word Embeddings

VIEW 3 EXCERPTS