Base language models outperform post-trained models at generating individual survey responses that match human distributions, while post-trained models are more accurate when directly predicting population opinion distributions.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Emulate or Estimate? The Divergent Strengths of Base and Post-Trained Language Models for Opinion Simulation
Base language models outperform post-trained models at generating individual survey responses that match human distributions, while post-trained models are more accurate when directly predicting population opinion distributions.