A new training scheme, Supervised Distillation Steering, separates on-policy teacher matching from adaptively-weighted canonical-context supervision, improving the plasticity-stability trade-off in LLM self-distillation.
Findings of the Association for Computational Linguistics: ACL 2023 , pages =
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Adaptive Supervised Anchoring for On-Policy Self-Distillation
A new training scheme, Supervised Distillation Steering, separates on-policy teacher matching from adaptively-weighted canonical-context supervision, improving the plasticity-stability trade-off in LLM self-distillation.