Removing the two deepest interaction blocks from the JMP-L molecular foundation model reduces size by 32% and speeds inference by 1.3x while keeping downstream accuracy nearly unchanged.
What is the state of neural network pruning? Proceedings of machine learning and systems , 2: 129–146, 2020
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
-
Towards Faster and More Compact Foundation Models for Molecular Property Prediction
Removing the two deepest interaction blocks from the JMP-L molecular foundation model reduces size by 32% and speeds inference by 1.3x while keeping downstream accuracy nearly unchanged.