Research map / Probabilistic and scientific modeling / Molecules and proteins
Protein Language Model Fitness is a Matter of Preference
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Studies when protein language models succeed or fail at predicting protein fitness, arguing that pLM fitness prediction depends on the preference of the sequence (fitness-related) distribution.
Problem
It is unclear under what conditions protein language models (pLMs) succeed or fail at aiding protein engineering via fitness prediction.
Approach
Frames pLM fitness prediction in terms of a model preference for sequences, aiming to predict the circumstances under which pLMs will succeed; the abstract does not specify the method in detail.
Closest papers by meaning
- DePLM: Denoising Protein Language Models for Property Optimization (NeurIPS 2024 · similarity 0.91)
- Steering Generative Models with Experimental Data for Protein Fitness Optimization (NeurIPS 2025 · similarity 0.90)
- Metalic: Meta-Learning In-Context with Protein Language Models (ICLR 2025 · similarity 0.90)
- Training Compute-Optimal Protein Language Models (NeurIPS 2024 · similarity 0.89)
- Steering Protein Language Models (ICML 2025 · similarity 0.89)
- Feature Reuse and Scaling: Understanding Transfer Learning with Protein Language Models (ICML 2024 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.