Orthogonal / Double Machine Learning: highly imbalanced treatment labels and highly skewed outcome labels?

Hi! Anyone knows of existing papers or methods, that applies Double ML to highly imbalanced dataset with, for example, a dataset with less than 1% samples being positively treated while the other >99% samples are untreated? Another example would be, a dataset with continuous outcome yet most Y has value being 0 and less than 1% outcome has great-than-zero value of Y?

Traditionally with supervised machine learning problems such as fraud detection, people use many techniques such as downsample/upsample the outcome, etc., to prevent the fitting of a trivial predictive model that predict every sample with the majority class. However this might introduce bias to Double ML's result coefficients in the last stage if we does this upsample / downsample in the first-stage ML models that predict T & Y. Any existing papers that have address this?

Start with the issue's questions about Double ML under highly imbalanced treatment and outcome labels, then review the EconML Python SDK's existing orthogonal machine-learning documentation. Done would require identifying whether established methods or papers address these cases and documenting guidance without introducing bias from first-stage resampling.