Full downstream results after standard 4-stage post-training pipeline (SFT, DPO, RLVR1, RLVR2) for NTP post-training baseline performance.

Full downstream results after standard 4-stage post-training pipeline (SFT, DPO, RLVR1, RLVR2) for NTP post-training baseline performance. · CiteArk

已评估 0/16 项

获得支持 0有争议或冲突 0尚未评估 16

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。
论文报告	最近可用实测	证据判断
67 percentage_points	—	尚未评估
46.8 percentage_points	—	尚未评估
40.6 percentage_points	—	尚未评估
31.1 percentage_points	—	尚未评估
32.7 percentage_points	—	尚未评估
12.6 percentage_points	—	尚未评估
53.3 percentage_points	—	尚未评估
21.6 percentage_points	—	尚未评估
7.5 percentage_points	—	尚未评估
32.3 percentage_points	—	尚未评估
16.6 percentage_points	—	尚未评估
52.4 percentage_points	—	尚未评估

报告指标与观测值