Tag
This paper investigates how epistemic stance (qualifiers, attributions) survives memory compression in AI agent memory systems. It finds that making the stance explicit as a labelled field improves retention significantly, while merely lengthening the text does not.
This paper introduces Epistemic Stance Flexibility Probing (ESFP), a behavioral benchmark that measures how large language models shift their epistemic register between attributing claims to experts and expressing their own stance. Evaluating eight frontier models, the authors find that epistemic flexibility is largely orthogonal to general capability, with stance content density providing the strongest signal.
Introduces a framework to quantify how LLMs overstate certainty through rhetorical devices, revealing model-agnostic patterns of epistemic-rhetorical miscalibration.