top of page
Wireframe - 8.png

Toward a Graded Measure of Belief Stability in Large Language Models

Samantha Dies, Branden Fitelson, Tina Eliassi-Rad

Working Paper

|

September 28, 2026

Large language models (LLMs) increasingly mediate how people access and reason with information, yet factual reliability is usually evaluated one judgment at a time. We introduce graded belief stability, a relational measure of how well a belief persists within an LLM's broader belief system. Unlike individual belief probability, it asks whether support for a claim persists when that claim is considered alongside the model's other epistemic commitments. We operationalize this idea with a Direct Conditional estimator that uses internal model representations to estimate conditional belief probabilities. Across 12 LLMs and three domains, lower-stability beliefs exhibit greater mean behavioral movement under conversational challenge in 83.3% of model-domain settings after matching on individual belief probability. Graded belief stability therefore extends reliability assessment beyond how strongly an LLM supports a claim to how robustly that belief is supported within its broader system of beliefs.

citation

Dies, S., Fitelson, B., & Eliassi-Rad, T. (2026). Toward a Graded Measure of Belief Stability in Large Language Models [Preprint]. arXiv. https://doi.org/10.48550/arxiv.2609.34158

topics

Artificial Intelligence

Artificial Intelligence

Artificial Intelligence

Artificial Intelligence

related work

Rectangle 35

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

Rectangle 35

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

Rectangle 35

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

bottom of page