A two-channel Allen-Dynes framework delivers blind Tc predictions with R²=0.96 for 19 materials over 0.4-250 K using phonon and spin-fluctuation channels extracted from data, while showing that the Peotta-Torma geometric superfluid weight correlates with topology rather than pairing strength.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2representative citing papers
pDOS-augmented GNN node features reduce Tc and ε∞ prediction errors by 22.9% and 27.9% versus elemental baselines, matching roughly a 1.7–1.8× increase in training data.
citing papers explorer
-
Two-Channel Allen-Dynes Framework for Superconducting Critical Temperatures: Blind Predictions Across Five Orders of Magnitude and a Quantum-Metric No-Go Result
A two-channel Allen-Dynes framework delivers blind Tc predictions with R²=0.96 for 19 materials over 0.4-250 K using phonon and spin-fluctuation channels extracted from data, while showing that the Peotta-Torma geometric superfluid weight correlates with topology rather than pairing strength.
-
Machine Learning Materials Properties by Encoding Orbital-Projected Density of States
pDOS-augmented GNN node features reduce Tc and ε∞ prediction errors by 22.9% and 27.9% versus elemental baselines, matching roughly a 1.7–1.8× increase in training data.