DePLM: Denoising Protein Language Models for Property Optimization Zeyuan Wang 1,2 Keyan Ding 2 Ming Qin 1,2 Xiaotong Li1,2

Protein optimization is a fundamental biological task aimed at enhancing the performance of proteins by modifying their sequences. Computational methods primarily rely on evolutionary information (EI) encoded by protein language models (PLMs) to predict fitness landscape for optimization. However, these methods suffer from a few limitations.