SIGIRMay, 2021
并非所有相关分数均相等:深度检索模型的高效不确定性和校准建模
Not All Relevance Scores are Equal: Efficient Uncertainty and Calibration Modeling for Deep Retrieval Models
Daniel Cohen, Bhaskar Mitra, Oleg Lesota, Navid Rekabsaz, Carsten Eickhoff
TL;DR以贝叶斯框架为基础的检索模型不仅能够提高排名的准确性,还能提供可靠的不确定性信息来预测截断点并提高下游任务的效果。