AI News/Wire summary

Science fiction shows more gender permeability than general fiction, study finds

Key points

  • Study masks man, woman and person in same-period science fiction and general fiction corpora, then classifies RoBERTa's five likeliest substitutes by gender.
  • Where woman is masked, the model predicts a masculine term about twice as often as a feminine term where man is masked.
  • Science fiction places man in feminine contexts more often than general fiction, and pulls both gendered terms toward neutral.
  • A masked woman draws social-category substitutes far more often than a masked man, according to the paper's findings.

Why it matters for marketers: Content teams using language models for gendered wording should expect asymmetric substitutions, and checks for woman versus man need separate review.

Summary by the Z Media news desk, from the outlets' reporting. Read the full stories at the source.

Read the original

arXiv cs.CL: Crossing the Gender Binary: Modelling Gendered Reference in Science Fiction and General Fiction

Related on Z Media