C. Danforth

Localized, unstable, and entangled: Exploring `Us vs. Them'bias in Large Language Models

Large language models (LLMs) exhibit us-vs.-them bias: A behavioral asymmetry in which prompts framed around an ingroup (``we''/``us'') receive systematically more positive continuations than matched prompts framed around an outgroup (``they''/``them''). Using Edge Attribution Patching (EAP), we localize this behavior...

Illness Signatures from Consumer Rings: Temperature, Respiration, Heart Rate, and Activity in a University Cohort

Self-reported archetypes and behavioral failures in Large Language Models

This work map the self-reported personality archetypes of 22 LLMs spanning closed-source frontier systems and open-source models, providing a reproducible, character-grounded framework for evaluating what LLMs are, not just what they do.

Archetypometrics of'Friends'

Storytelling inherently revolves around characters. Using the television sitcom `Friends'as a case study, we investigate how well archetype vectors capture both individual characterization and the relational structure of a specific ensemble. Our work is based on the archetypometrics framework, which locates 2,000 ficti...