Emulate or Estimate? The Divergent Strengths of Base and Post-Trained Language Models for Opinion Simulation

• URL: http://arxiv.org/abs/2608.03044v1
• Date: Tue, 04 Aug 2026 02:51:38 GMT
• Title: Emulate or Estimate? The Divergent Strengths of Base and Post-Trained Language Models for Opinion Simulation
• Abstract summary: We show that some studies find promising alignment with human survey data, while others find persona collapse and weak demographic sensitivity.<n>We call the first task emulation, in which models generate individual responses that aggregate into a population distribution.<n>We call the second task estimation, in which models directly predict the population distribution.
• Score: 7.947622351777421
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large language models are increasingly used to simulate human opinions, but prior work reports conflicting results: some studies find promising alignment with human survey data, while others find persona collapse and weak demographic sensitivity. We show that much of this conflict stems from conflating two distinct tasks. We call the first task emulation, in which models generate individual responses that aggregate into a population distribution. We call the second task estimation, in which models directly predict the population distribution. Evaluating six matched base and post-trained models on the Pew American Trends Panel, we find that base models are stronger emulators: they produce response distributions closer to human ground truth and better preserve demographic structure. Post-trained models are stronger estimators, producing more accurate distributional predictions when asked directly. We propose that model selection for human simulation should be guided by whether the task requires generating text or predicting distributions.

Related papers

• OdysSim: Building Foundation Models for Human Behavior Simulation [70.35265860287608]
  Large language models are increasingly deployed as human simulators for interactive evaluation and social simulation.<n>Yet helpfulness-driven post-training pulls them toward a homogeneous, overly agreeable assistant register.<n>We present OdysSim, the largest open systematic investigation of behavioral foundation models.
  arXiv  Detail & Related papers  (2026-06-12T07:31:55Z)
• Shape of Thought: When Distribution Matters More than Correctness in Reasoning Tasks [24.55929874173401]
  We show that a language model's reasoning capabilities can be improved by training on datasets of chain-of-thought traces from more capable models.<n>Experiments show this approach can yield better performance on reasoning tasks than training on human-annotated datasets.
  arXiv  Detail & Related papers  (2025-12-24T07:35:55Z)
• Specializing Large Language Models to Simulate Survey Response Distributions for Global Populations [49.908708778200115]
  We are the first to specialize large language models (LLMs) for simulating survey response distributions.<n>As a testbed, we use country-level results from two global cultural surveys.<n>We devise a fine-tuning method based on first-token probabilities to minimize divergence between predicted and actual response distributions.
  arXiv  Detail & Related papers  (2025-02-10T21:59:27Z)
• Evaluation of Large Language Models via Coupled Token Generation [19.187846871216568]
  State of the art large language models rely on randomization to respond to a prompt.<n>We argue that the evaluation and ranking of large language models should control for the randomization underpinning their functioning.
  arXiv  Detail & Related papers  (2025-02-03T19:01:17Z)
• Using LLMs to Model the Beliefs and Preferences of Targeted Populations [4.0849074543032105]
  We consider the problem of aligning a large language model (LLM) to model the preferences of a human population. Modeling the beliefs, preferences, and behaviors of a specific population can be useful for a variety of different applications.
  arXiv  Detail & Related papers  (2024-03-29T15:58:46Z)
• Random Silicon Sampling: Simulating Human Sub-Population Opinion Using a Large Language Model Based on Group-Level Demographic Information [15.435605802794408]
  Large language models exhibit societal biases associated with demographic information. We propose "random silicon sampling," a method to emulate the opinions of the human population sub-group. We find that language models can generate response distributions remarkably similar to the actual U.S. public opinion polls.
  arXiv  Detail & Related papers  (2024-02-28T08:09:14Z)
• Beyond Human Data: Scaling Self-Training for Problem-Solving with Language Models [115.501751261878]
  Fine-tuning language models(LMs) on human-generated data remains a prevalent practice. We investigate whether we can go beyond human data on tasks where we have access to scalar feedback. We find that ReST$EM$ scales favorably with model size and significantly surpasses fine-tuning only on human data.
  arXiv  Detail & Related papers  (2023-12-11T18:17:43Z)
• Questioning the Survey Responses of Large Language Models [25.14481433176348]
  We critically examine the methodology on the basis of the well-established American Community Survey by the U.S. Census Bureau.<n>We establish two dominant patterns. First, models' responses are governed by ordering and labeling biases, for example, towards survey responses labeled with the letter "A"<n>Second, when adjusting for these systematic biases through randomized answer ordering, models across the board trend towards uniformly random survey responses.
  arXiv  Detail & Related papers  (2023-06-13T17:48:27Z)
• How to Estimate Model Transferability of Pre-Trained Speech Models? [84.11085139766108]
  "Score-based assessment" framework for estimating transferability of pre-trained speech models. We leverage upon two representation theories, Bayesian likelihood estimation and optimal transport, to generate rank scores for the PSM candidates. Our framework efficiently computes transferability scores without actual fine-tuning of candidate models or layers.
  arXiv  Detail & Related papers  (2023-06-01T04:52:26Z)
• Transformer-Based Language Model Surprisal Predicts Human Reading Times Best with About Two Billion Training Tokens [17.80735287413141]
  We evaluate surprisal estimates from Transformer-based language model variants on their ability to predict human reading times. Results show that surprisal estimates from most variants with contemporary model capacities provide the best fit after seeing about two billion training tokens. Newly-trained smaller model variants reveal a 'tipping point' at convergence, after which the decrease in language model perplexity begins to result in poorer fits to human reading times.
  arXiv  Detail & Related papers  (2023-04-22T12:50:49Z)
• Dialogue Response Ranking Training with Large-Scale Human Feedback Data [52.12342165926226]
  We leverage social media feedback data to build a large-scale training dataset for feedback prediction. We trained DialogRPT, a set of GPT-2 based models on 133M pairs of human feedback data. Our ranker outperforms the conventional dialog perplexity baseline with a large margin on predicting Reddit feedback.
  arXiv  Detail & Related papers  (2020-09-15T10:50:05Z)
• Are Visual Explanations Useful? A Case Study in Model-in-the-Loop Prediction [49.254162397086006]
  We study explanations based on visual saliency in an image-based age prediction task. We find that presenting model predictions improves human accuracy. However, explanations of various kinds fail to significantly alter human accuracy or trust in the model.
  arXiv  Detail & Related papers  (2020-07-23T20:39:40Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.