Exoformer uses a transformer network to generate informative priors that accelerate Bayesian atmospheric retrievals of hot Jupiters by 3-8 times without altering the final parameters or Bayesian evidence.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
astro-ph.EP 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
$\texttt{Exoformer}$: Accelerating Bayesian atmospheric retrievals with transformer neural networks
Exoformer uses a transformer network to generate informative priors that accelerate Bayesian atmospheric retrievals of hot Jupiters by 3-8 times without altering the final parameters or Bayesian evidence.