Search papers, labs, and topics across Lattice.
This paper investigates how personality traits influence gender bias in persona-conditioned LLM story generation across English and Hindi. By generating 23,400 stories from six LLMs with systematically varied persona gender, occupation, and personality traits (HEXACO and Dark Triad), the study reveals that Dark Triad traits correlate with increased gender-stereotypical representations compared to HEXACO traits. The findings highlight the context-dependent nature of gender bias in LLMs, suggesting that persona conditioning can exacerbate representational harms.
LLMs' gender biases aren't fixed; they warp and intensify based on the *personality* you give them, especially when those personalities lean toward the "Dark Triad."
Large Language Models (LLMs) are increasingly deployed in persona-driven applications such as education, customer service, and social platforms, where models are prompted to adopt specific personas when interacting with users. While persona conditioning can improve user experience and engagement, it also raises concerns about how personality cues may interact with gender biases and stereotypes. In this work, we present a controlled study of persona-conditioned story generation in English and Hindi, where each story portrays a working professional in India producing context-specific artifacts (e.g., lesson plans, reports, letters) under systematically varied persona gender, occupational role, and personality traits from the HEXACO and Dark Triad frameworks. Across 23,400 generated stories from six state-of-the-art LLMs, we find that personality traits are significantly associated with both the magnitude and direction of gender bias. In particular, Dark Triad personality traits are consistently associated with higher gender-stereotypical representations compared to socially desirable HEXACO traits, though these associations vary across models and languages. Our findings demonstrate that gender bias in LLMs is not static but context-dependent. This suggests that persona-conditioned systems used in real-world applications may introduce uneven representational harms, reinforcing gender stereotypes in generated educational, professional, or social content.
