Pre-training improves CLMs' encoding of molecular substructures especially in upper layers while fine-tuning selectively modifies task-relevant substructures more than others.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Probing Chemical Language Models: Effects of Pre-training and Fine-tuning
Pre-training improves CLMs' encoding of molecular substructures especially in upper layers while fine-tuning selectively modifies task-relevant substructures more than others.