Re-retrieval improves QA accuracy for retrieval failures but not extraction failures, and corpus-wide answer-string searches report that the required relational fact is absent in nearly all extraction-failure cases across all three datasets.

Re-retrieval improves QA accuracy for retrieval failures but not extraction failures, and corpus-wide answer-string searches report that the required relational fact is absent in nearly all extraction-failure cases across all three datasets. · CiteArk

测量结果

已评估 0/6 项

获得支持 0有争议或冲突 0尚未评估 6

保留同一结论下的完整结果组。已有评估排在前面；单项判断不替代整组结论。展开行查看条件、来源与历次证据。

每行是一项测量，分别显示论文报告、最近可用实测和累计证据状态。
指标与条件	论文报告	最近可用实测	证据判断
intervention: augmentation	8.6 percentage points	—	尚未评估
intervention: reranking	25.3 percentage points	—	尚未评估
intervention: re-retrieval; intervention subtype not separately reported for this value	-0.5 percentage points	—	尚未评估
dataset: MuSiQue	97.7%	—	尚未评估
dataset: HotpotQA	100%	—	尚未评估
dataset: 2WikiMultihopQA	97.6%	—	尚未评估

Re-retrieval improves QA accuracy for retrieval failures but not extraction failures, and corpus-wide answer-string searches report that the required relational fact is absent in nearly all extraction-failure cases across all three datasets.

报告指标与观测值

Assessment（0）

关联运行（0）

测量结果