Back to feed
arXiv cs.LG
arXiv cs.LG
8/5/2026
Multimodal Auto-regressive Transformer Surrogate for Modeling Variable Operations and Quantifying Uncertainty in Geological Carbon Storage

Multimodal Auto-regressive Transformer Surrogate for Modeling Variable Operations and Quantifying Uncertainty in Geological Carbon Storage

Short summary

Researchers develop a multimodal auto-regressive transformer surrogate for modeling variable well operations in geological carbon storage. The model fuses 3D geomodels, scalar permeability parameters, and control variables via self-attention, trained on 4000 GEOS flow simulations. It achieves median saturation MAE of 0.028 and enables hierarchical MCMC data assimilation with substantial uncertainty reduction for fault permeabilities.

  • •Multimodal transformer surrogate models geological carbon storage operations
  • •Trained on 4000 simulations; achieves low MAE for saturation predictions
  • •Enables MCMC-based uncertainty reduction for key geological metaparameters

Generated with AI, which can make mistakes.

#research-breakthrough
Read full article at arXiv cs.LG

Is this a good recommendation for you?

Comments

Failed to load comments. Please try again.

Explore more

Towards Fast GNN Surrogates for CO2 Migration in Complex Geological Formations
arXiv cs.LG
Uncertainty-Aware Simulation-Based Inference for Operations Research with Large Language Models
arXiv cs.LG
BattVAE-GP: Generative Modeling of Long-Horizon Battery Degradation with Uncertainty Quantification
arXiv cs.LG
Mind the Residual Gap: Probabilistic Downscaling under Real-World Bias
arXiv cs.LG