10.1101/2025.05.29.656902
Trainable subnetworks reveal insights into structure knowledge organization in protein language models
2025-06-01