论文Scaling Up Crowd-Sourcing to Very Large Datasets A Case for Active Learning提出两种AL算法。   首先找到分类器θ对未标注数据的不确定程度。然后让crowd对这些数据进行标定。下边介绍两种不确定性方法。 下边的u是未标记数据，但是是指未标注数据的每一个，而不是整体。 一：Uncertainty Algorithm

Retrieve & Re-Rank https://www.sbert.net/examples/applications/retrieve_rerank/README.html Bi-Encoder vs. Cross-Encoder https://www.sbert.net/examples/applications/cross-encoder/README.html Bi-Encode