llm social bias
Research on bias, homogenization, mode collapse, and diversity in large language models. Critical and queer-theoretic approaches to AI safety.
-
Proposes a pipeline that discovers and measures how deployed LLMs treat queer and cisheterosexual users differently, without relying on explicit markers
-
Foregrounds homogenization as a central AI safety concern; uses critical and queer theory to formalize diversity, normativity, and xeno-reproduction in LLMs
-
Characterizes social bias in Spanish-prompted LLMs across model sizes with the SESGO benchmark; smaller models show more bias and yield more to prompt scaffolds
-
Formalizing xeno-reproduction as structure-aware diversity pursuit to mitigate homogenization in generative AI
-
Exploring diversity and alternative modes of reproduction as an AI safety objective
