Annotation Efficiency: Identifying Hard Samples via Blocked Sparse Linear Bandits

Open in new window