few-shot anomaly detection (AD) is an emerging sub-field of general AD, and
tries to distinguish between normal and anomalous data using only few selected
samples. While newly proposed few-shot AD methods do compare against
pre-existing algorithms developed for the full-shot domain as