An evaluation of eight LLMs on a hand-built risk-tolerance rubric finds large scoring errors and small demographic differences, but the rubric given to the models contradicts the rubric used as ground truth.
Nucleation and growth manifest universal scaling, surely
1 Pith paper cite this work. Polarity classification is still indexing.
abstract
When a system is brought to a metastable state, nuclei of the equilibrium phase form and grow. This is the well-known nucleation and growth of first-order phase transitions. Near a critical point of a continuous phase transition, critical phenomena such as critical opalescence characterized by universal scaling emerge. These two sets of behavior are so completely different that it might appear absurd to ask whether nucleation and growth exhibit even a slight universal scaling. Here we show that universal scaling is indeed manifested in standard nucleation and growth by introducing a non-universal initial time to a recently proposed method. This generalizes the method and provides a different perspective for studying universal scaling behavior and hence universality classes of nucleation and growth.
fields
cs.CL 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Evaluating AI for Finance: Is AI Credible at Assessing Investment Risk?
An evaluation of eight LLMs on a hand-built risk-tolerance rubric finds large scoring errors and small demographic differences, but the rubric given to the models contradicts the rubric used as ground truth.