Papers by Maria Symeonaki
Assumed Identities: Quantifying Gender Bias in Machine Translation of Gender-Ambiguous Occupational Terms (2025.emnlp-main)
Copied to clipboard
| Challenge: | ailsntua researchers examine whether machine translation systems exhibit gender biases that reinforce societal stereotypes. |
| Approach: | They propose a probability-based metric to evaluate gender bias by analyzing aggregated model responses. |
| Outcome: | The proposed metric evaluates whether translations in Greek and French align with or diverge from societal stereotypes. |