1
Return

Machine learning thermobarometry: Methods, applications, and a benchmarking protocol

delete2026-06-15
delete0
PRE
AI
X
Xirui Qiao
Y
Yan Xia *
X
Xisheng Xu
DOI：10.1016/j.lithos.2026.108665delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
• A Modular Benchmarking Protocol (MBP) is proposed for ML thermobarometry. • Coexisting melt compositions reduce T/P RMSE by 40.9%/16.1% vs. NoLiquid baseline. • Physics-informed data augmentation significantly enhances model robustness. • Predictive precision approaches the analytical limit. • Input information density is the primary bottleneck for thermobarometric accuracy.

Journal

Lithos cover
Lithos
IF:
2.5
Papers:
649
Citations:
2.8W

Organization

No organization information available
Cited Papers

Cited Papers

Citing Papers

Citing Papers