Papers by Mae Sosto
QueerGen: How LLMs Reflect Societal Norms on Gender and Sexuality in Sentence Completion Task (2026.findings-eacl)
Copied to clipboard
| Challenge: | Autoregressive Language Models (ARLMs) partially mitigate these patterns, while closed-access ARLMs tend to produce more harmful outputs for unmarked subjects. |
| Approach: | They examine whether explicit information about a subject’s gender or sexuality influences LLM responses across three subject categories: queer-marked, non-queer-mark, and the normalized "unmarked" category. |
| Outcome: | The proposed models reproduce normative social assumptions, but the form and degree of bias depend on model characteristics, which may redistribute—but not eliminate—representational harms. |