Linear Fitness Subspace in Protein Language Models Enables Sample-Efficient Directed Evolution

Excerpt

arXiv:2610.07607v1 Announce Type: cross Abstract: Model-guided directed evolution seeks to identify high-fitness protein variants under limited oracle budgets. Protein language models (PLMs) provide rich representations for this task, but task-agnostic zero-shot scores can be misaligned with a target assay, while supervised search in high-dimensional embedding spaces can make surrogate modeling and uncertainty estimation sample-inefficient. We propose the Linear Fitness Subspace (LFS) hypothesis