Fetching the paper…

Attention in Large Language Models Yields Efficient Zero-Shot Re-Rankers · Around