Growth-based pretraining (StackLLM) forgets less in continual fine-tuning, especially on reading comprehension, but reduces social bias less than standard training.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Mitigating Catastrophic Forgetting in Continual Learning through Model Growth
Growth-based pretraining (StackLLM) forgets less in continual fine-tuning, especially on reading comprehension, but reduces social bias less than standard training.