Prompts to Proxies: Emulating Human Preferences via a Compact LLM Ensemble
cs.AI, cs.CY
Submitted: 2025-09-14
Updated: 2026-10-27
Comments: Accepted for publication at NeurIPS 2026
License: http://creativecommons.org/licenses/by/4.0/
The gist: Large language models are increasingly used as proxies for human subjects in social science research, yet external validity requires matching the response distributions of target human populations.
Terminology
Abstract
Large language models are increasingly used as proxies for human subjects in social science research, yet external validity requires matching the response distributions of target human populations. We study population-level survey alignment: reconstructing aggregate survey responses from limited public survey data, without individual-level demographic profiles or model finetuning. We formalize this problem as preference reconstruction: rather than matching proxy agents to demographic profiles, we construct a functional basis of proxy agents and recover population preferences by weighted aggregation. We instantiate this idea via Prompts to Proxies (P2P), a two-stage inference-time system. Stage 1 uses structured attribute-based prompting with entropy-guided adaptive sampling to construct a diverse proxy pool spanning the latent preference space. Stage 2 employs L1-regularized regression to select a compact weighted ensemble matching observed target-population responses. Across 14 American Trends Panel waves, P2P achieves an average test MSE of 0.014 at approximately 0.8 USD per survey, improving over prompting and demographic-conditioning baselines. On the World Values Survey, cross-locale transfer experiments show that basis expressiveness can outperform locale matching on average, while locale-specific generation helps on culturally divergent questions. A stress test against an SFT-aligned survey model shows competitive performance using less than 3% of the training data. These results position preference reconstruction as a lightweight, externally verifiable alternative for survey-based population alignment. Code is available at https://github.com/bingchen-wang/P2P.
Sources
- Steerable Pluralism: Pluralistic Alignment via Few-Shot Comparative Regression
- Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback
- SPICA: Retrieving Scenarios for Pluralistic In-Context Alignment
- Towards Measuring the Representation of Subjective Global Opinions in Language Models
- Policy Prototyping for LLMs: Pluralistic Alignment via Interactive and Collaborative Policymaking
- AI Alignment: A Comprehensive Survey
- AI Simulations Grounded in Rich Self-Reports Increase Accuracy and Reduce Distortion
- Cultural Palette: Pluralising Culture Alignment via Multi-agent Palette
- Fair-PP: A Synthetic Dataset for Aligning LLM with Personalized Preferences of Social Equity
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents