Fine-tuning LLMs on small pilot survey data balances structural, marginal, and individual fidelity better than prompting or rectification, but fidelity levels vary across subsamples in a COVID-19 misinformation case study.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CL 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
SLM responses to psychometric prompts are dominated by prompt artifacts such as personas and option symbols rather than semantic understanding of psychological constructs.
citing papers explorer
-
Beyond the Mean: Three-Axis Fidelity for Aligning LLM-Based Survey Simulators from Small Pilot Data
Fine-tuning LLMs on small pilot survey data balances structural, marginal, and individual fidelity better than prompting or rectification, but fidelity levels vary across subsamples in a COVID-19 misinformation case study.
-
The Unsampled Truth: Psychometrics in SLMs Measure Prompt Artifacts, Not Psychological Constructs
SLM responses to psychometric prompts are dominated by prompt artifacts such as personas and option symbols rather than semantic understanding of psychological constructs.