A subsequence-reordering pretraining objective with variable-length protein cuts improves zero-shot compound-protein interaction prediction and is data-efficient relative to large protein language models.
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
-
Zero-Shot Learning with Subsequence Reordering Pretraining for Compound-Protein Interaction
A subsequence-reordering pretraining objective with variable-length protein cuts improves zero-shot compound-protein interaction prediction and is data-efficient relative to large protein language models.