Apparent psychological profiles of LLMs are largely measurement artifacts driven by directional response bias rather than actual traits.
Mixed citations
Bennett, Shaun K
Mixed citation behavior. Most common role is background (60%).
7
Pith papers citing it
1
external citations · external index
Background
60% of classified citations
citation-role summary
background 4
dataset 1
citation-polarity summary
years
2026 7representative citing papers
LLM self-reports predict behavior selectively: TPB reaches human-level coherence within shared conversations but collapses across sessions for primed behaviors, unlike Big 5, with persona prompting stabilizing reports but not actions.
Thematic analysis of 43 AI contestation cases, using Bovens's relational accountability model, produces categories of demands from below, institutional pushback, outcomes, and contextual factors shaping effective contestation.
Automated hate speech detectors show poor alignment with heterogeneous in-group judgments on reclaimed slur usage, driven by low inter-annotator agreement and contextual features like derogatory intent.
Case studies with blind UK residents and people from Kerala and Tamil Nadu demonstrate that community input at the systematization stage produces culturally grounded definitions of appropriateness for text-to-image model outputs.
A literature review concludes that pursuing consensus in data annotation creates biased AI by dismissing subjective disagreements and enforcing geographic hegemony, and proposes mapping diversity instead.
citing papers explorer
-
Apparent Psychological Profiles of Large Language Models are Largely a Measurement Artifact
Apparent psychological profiles of LLMs are largely measurement artifacts driven by directional response bias rather than actual traits.
-
Rethinking Psychometric Evaluation of LLMs: When and Why Self-Reports Predict Behavior
LLM self-reports predict behavior selectively: TPB reaches human-level coherence within shared conversations but collapses across sessions for primed behaviors, unlike Big 5, with persona prompting stabilizing reports but not actions.
-
Push and Pushback in Contesting AI: Demands for and Resistance to Accountability
Thematic analysis of 43 AI contestation cases, using Bovens's relational accountability model, produces categories of demands from below, institutional pushback, outcomes, and contextual factors shaping effective contestation.
-
IYKYK (But AI Doesn't): Automated Content Moderation Does Not Capture Communities' Heterogeneous Attitudes Towards Reclaimed Language
Automated hate speech detectors show poor alignment with heterogeneous in-group judgments on reclaimed slur usage, driven by low inter-annotator agreement and contextual features like derogatory intent.
-
Evaluating AI-Generated Images of Cultural Artifacts with Community-Informed Rubrics
Case studies with blind UK residents and people from Kerala and Tamil Nadu demonstrate that community input at the systematization stage produces culturally grounded definitions of appropriateness for text-to-image model outputs.
-
The Consensus Trap: Dissecting Subjectivity and the "Ground Truth" Illusion in Data Annotation
A literature review concludes that pursuing consensus in data annotation creates biased AI by dismissing subjective disagreements and enforcing geographic hegemony, and proposes mapping diversity instead.
- Model Multiplicity and Predictive Arbitrariness in Recidivism Risk Assessment