Calibrating the teacher with temperature scaling before knowledge distillation gives small student accuracy gains, but the paper's causal claim that calibration error is the key driver is not cleanly established.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
The Role of Teacher Calibration in Knowledge Distillation
Calibrating the teacher with temperature scaling before knowledge distillation gives small student accuracy gains, but the paper's causal claim that calibration error is the key driver is not cleanly established.