<ul><li>Retrieval augmentation, the practice of retrieving additional data from large auxiliary pools, has emerged as an effective technique for enhancing model performance in the low-data regime.</li><li>Prior approaches have employed only nearest-neighbor based strategies for data selection, which retrieve auxiliary samples with high similarity to instances in the target task.</li><li>COBRA (COmBinatorial Retrieval Augmentation) is a new approach that employs an alternative CMI measure that considers both diversity and similarity to a target dataset for retrieval augmentation.</li><li>COBRA consistently outperforms previous retrieval approaches, providing significant gains in downstream model performance without incurring significant computational overhead.</li></ul>

COBRA: COmBinatorial Retrieval Augmentation for Few-Shot Adaptation

Discover more