Diffusion models trained on Monte Carlo Ising configurations reproduce energy, magnetization, and general fluctuation trends across the phase transition, with a distorted critical exponent and low-temperature specific heat; GANs need physics-informed loss terms to partially match.
Distribution of potential energies for (a) Monte Carlo, (b) DiffIsing2, and (c) DiffFlip models
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cond-mat.stat-mech 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Thermodynamic Fidelity of Generative Models for Ising System
Diffusion models trained on Monte Carlo Ising configurations reproduce energy, magnetization, and general fluctuation trends across the phase transition, with a distorted critical exponent and low-temperature specific heat; GANs need physics-informed loss terms to partially match.