Reading Between the Tokens: Improving Preference Predictions through Mechanistic Forecasting

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Ball, Sarah, Allmendinger, Simeon, Kühl, Niklas, Kreuter, Frauke
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915770245578752
author Ball, Sarah
Allmendinger, Simeon
Kühl, Niklas
Kreuter, Frauke
author_facet Ball, Sarah
Allmendinger, Simeon
Kühl, Niklas
Kreuter, Frauke
contents Large language models are increasingly used to predict human preferences in both scientific and business endeavors, yet current approaches rely exclusively on analyzing model outputs without considering the underlying mechanisms. Using election forecasting as a test case, we introduce mechanistic forecasting, a method that demonstrates that probing internal model representations offers a fundamentally different - and sometimes more effective - approach to preference prediction. Examining over 24 million configurations across 7 models, 6 national elections, multiple persona attributes, and prompt variations, we systematically analyze how demographic and ideological information activates latent party-encoding components within the respective models. We find that leveraging this internal knowledge via mechanistic forecasting (opposed to solely relying on surface-level predictions) can improve prediction accuracy. The effects vary across demographic versus opinion-based attributes, political parties, national contexts, and models. Our findings demonstrate that the latent representational structure of LLMs contains systematic, exploitable information about human preferences, establishing a new path for using language models in social science prediction tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2602_02882
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Reading Between the Tokens: Improving Preference Predictions through Mechanistic Forecasting
Ball, Sarah
Allmendinger, Simeon
Kühl, Niklas
Kreuter, Frauke
Computers and Society
Large language models are increasingly used to predict human preferences in both scientific and business endeavors, yet current approaches rely exclusively on analyzing model outputs without considering the underlying mechanisms. Using election forecasting as a test case, we introduce mechanistic forecasting, a method that demonstrates that probing internal model representations offers a fundamentally different - and sometimes more effective - approach to preference prediction. Examining over 24 million configurations across 7 models, 6 national elections, multiple persona attributes, and prompt variations, we systematically analyze how demographic and ideological information activates latent party-encoding components within the respective models. We find that leveraging this internal knowledge via mechanistic forecasting (opposed to solely relying on surface-level predictions) can improve prediction accuracy. The effects vary across demographic versus opinion-based attributes, political parties, national contexts, and models. Our findings demonstrate that the latent representational structure of LLMs contains systematic, exploitable information about human preferences, establishing a new path for using language models in social science prediction tasks.
title Reading Between the Tokens: Improving Preference Predictions through Mechanistic Forecasting
topic Computers and Society
url https://arxiv.org/abs/2602.02882