Publications and scientific references
These references do not exist to inflate claims. They help document the scientific context around synthetic populations, virtual users, LLM orchestration and the interpretation of simulated audience outputs.
How to read this list
Each reference matters because it clarifies a part of the stack: population construction, virtual-user simulation, language-model orchestration or environmental and governance framing. Together they give Pollitics a more legible scientific perimeter.
Why a structured publications page matters
A structured publications page makes it easier to understand which methods, papers, authors and institutions inform the product. It also clarifies how terms like synthetic panel, virtual users or AI qualitative research should be interpreted in context.
Scientific references
Explains a constraint-programming approach for generating exact synthetic populations from aggregated statistics, with direct relevance for coherent synthetic panels and privacy-preserving audience simulation.
LLMs, Virtual Users, and BiasExamines virtual users, survey prediction and bias, which is directly relevant when framing what simulated respondents can and cannot be used for.
A Declarative Framework for Large Language Model IntegrationDocuments how task-specific orchestration of language models can be made more explicit and controlled, which matters for robust simulation workflows.
Transparent Screening for LLM Inference and Training ImpactsAdds governance and transparency context around LLM usage, useful when positioning Pollitics as a serious, documented AI system rather than a black-box demo.
Exact Synthetic Populations for Scalable Societal and Market Modeling (CPAIOR-2026 proceedings)Conference version of the exact synthetic population work, listed in the accepted papers of CPAIOR 2026 and directly aligned with the population-construction layer behind Pollitics.
FAQ
Are all these publications directly product papers about Pollitics?
Not all of them. Some are directly tied to synthetic populations or virtual users, while others document adjacent methodological layers that shape the product.
Should prospects read the papers before using Pollitics?
No. The page is here to provide scientific traceability and context, not to turn product evaluation into an academic exercise.