The fact-grounding gap is small in aggregate but heterogeneous on 2WikiMultihopQA: extraction failures are rare for comparison questions and substantially more frequent for multi-step questions; domain-specific predictor retraining is accurate but still does not make re-retrieval beneficial.

The fact-grounding gap is small in aggregate but heterogeneous on 2WikiMultihopQA: extraction failures are rare for comparison questions and substantially more frequent for multi-step questions; domain-specific predictor retraining is accurate but still does not make re-retrieval beneficial. · CiteArk

测量结果

已评估 0/9 项

获得支持 0有争议或冲突 0尚未评估 9

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
questionType: comparison	2.7%	—	尚未评估
questionType: multi-step	16.6%	—	尚未评估
	97.4%	—	尚未评估
	0.983 score	—	尚未评估
	0%	—	尚未评估
predictor: 2WikiMultihopQA-retrained	78.9%	—	尚未评估
	78.8%	—	尚未评估
	0 probability	—	尚未评估
	0.999 probability	—	尚未评估

The fact-grounding gap is small in aggregate but heterogeneous on 2WikiMultihopQA: extraction failures are rare for comparison questions and substantially more frequent for multi-step questions; domain-specific predictor retraining is accurate but still does not make re-retrieval beneficial.

报告指标与观测值

Assessment（0）

关联运行（0）

测量结果