Knowledge Distillation During Mid-Training Favors Reasoning over Factual Recall
尚未评估计划受阻研究发现claim_fig5_gradient_attenuation_factual_supervision
Higher teacher entropy corresponds to strictly lower gold-token probability, attenuates the gold-token gradient relative to NTP (reaching ~0.5x NTP for Q5 facts), and produces larger downstream factual recall deficits.
来源：source_paper:Figure 5, Page 7
报告指标与观测值
这条 Claim 暂无结构化指标。
Assessment（0）
这条 Claim 暂无已发布的不可变 Assessment。
关联运行（0）
这条 Claim 暂无关联执行。