
Representational and Behavioral Stability of Truth in Large Language Models
Samantha Dies, Courtney Maynard, Germans Savcisens, Tina Eliassi-Rad
Working Paper
|
July 22, 2026
Large language models (LLMs) are widely used as information sources, yet small changes in semantic assumptions can destabilize their beliefs. We introduce P-StaT (Perturbation Stability of Truth), a framework for evaluating belief stability under matched semantic perturbations in both representational and behavioral settings. Across 21 LLMs and three domains, we compare perturbations involving familiar Fictional statements against synthetically generated and unfamiliar Synthetic statements. Unfamiliar Synthetic perturbations generally induce greater epistemic instability than familiar Fictional perturbations, with behavioral belief retraction rates frequently exceeding 0.5 (50%). Finally, exploratory clustering analyses reveal recurring themes among retracted statements, including ambiguity, technical terminology, and obscure concepts. These results show that epistemic familiarity is systematically associated with stability under semantic reframing, suggesting that stability-based analyses can complement accuracy-based benchmarks when evaluating LLM robustness and reliability. Code and data are available at https://github.com/samanthadies/P-StaT.
citation
Dies, S., Maynard, C., Savcisens, G., & Eliassi-Rad, T. (2026). Representational and Behavioral Stability of Truth in Large Language Models [Preprint]. SSRN. https://doi.org/10.2139/ssrn.6097886
topics
Artificial Intelligence
Artificial Intelligence
Artificial Intelligence
Artificial Intelligence
related work

Measuring Agency-Amplifying AI Literacy: The HuskyAI Coaching Platform for Workforce Readiness
July 2026
HuskyAI is a web platform on which students develop AI knowledge by working through authentic
Artificial Intelligence
Artificial Intelligence
Working Paper


Measuring Agency-Amplifying AI Literacy: The HuskyAI Coaching Platform for Workforce Readiness
July 2026
HuskyAI is a web platform on which students develop AI knowledge by working through authentic
Artificial Intelligence
Artificial Intelligence
Working Paper


Measuring Agency-Amplifying AI Literacy: The HuskyAI Coaching Platform for Workforce Readiness
July 2026
HuskyAI is a web platform on which students develop AI knowledge by working through authentic
Artificial Intelligence
Artificial Intelligence
Working Paper

.png)