Big Five inventories fail to capture meaningful differences or recover the five-factor structure in LLMs, with only 3% variance between models and four facets collapsing (r >= .92).
PNAS Nexus , volume =
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2representative citing papers
Big Five personality traits become decodable early in LLMs, are represented by mid-layer selective neurons, and can be shifted by targeted activation interventions, though effects on generated labels are weaker and spill across traits.
citing papers explorer
-
Personality Without Persons? A Psychometric Critique of Big Five Testing in Large Language Models
Big Five inventories fail to capture meaningful differences or recover the five-factor structure in LLMs, with only 3% variance between models and four facets collapsing (r >= .92).
-
Psychological Concept Neurons: Can Neural Control Bias Probing and Shift Generation in LLMs?
Big Five personality traits become decodable early in LLMs, are represented by mid-layer selective neurons, and can be shifted by targeted activation interventions, though effects on generated labels are weaker and spill across traits.