Knowledge Distillation During Mid-Training Favors Reasoning over Factual Recall
Not assessedPlan blockedMeasurementclaim_t3_random_routing
Mid-training ablation results for Random Routing relative difference across Reasoning, Factual Recall, and Knowledge task groups.
Source: source_paper:Table 3, Page 9
Reported and observed measurements
delta_percentage_points
t3_random_routing_reasoning
Reported -6.5 percentage_points
Observed — percentage_points
delta_percentage_points
t3_random_routing_factual_recall
Reported -0.8 percentage_points
Observed — percentage_points
delta_percentage_points
t3_random_routing_knowledge
Reported -2 percentage_points
Observed — percentage_points
Assessments (0)
No immutable Assessment has been published for this Claim yet.
Runs (0)
No execution has been linked to this Claim yet.