All five tested OpenAI LLMs show significant gender bias in occupational predictions, aligning more with U.S. labor statistics than with a neutral benchmark or human perception.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2024 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Assessing Gender Bias in LLMs: Comparing LLM Outputs with Human Perceptions and Official Statistics
All five tested OpenAI LLMs show significant gender bias in occupational predictions, aligning more with U.S. labor statistics than with a neutral benchmark or human perception.