A single diffusion model with a power-law noise schedule can generate seismic data and perform posterior sampling for several noisy linear inverse problems without task-specific retraining.
Frequency-dependent avo inversion and application on tight sandstone gas reservoir prediction using deep neural network
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
physics.geo-ph 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Generative modeling of seismic data using diffusion models and its application to multi-purpose posterior sampling for noisy inverse problems
A single diffusion model with a power-law noise schedule can generate seismic data and perform posterior sampling for several noisy linear inverse problems without task-specific retraining.