# Repurposing Protein Language Models for Latent Flow-Based Fitness Optimization

Preprint · Amaru Caceres Arroyo, Lea Bogensperger, Ahmed Allam, Michael Krauthammer, Konrad Schindler, Dominik Narnhofer

- Archive id: arxiv:2602.02425
- Published: 2026-02-02
- arXiv: https://arxiv.org/abs/2602.02425
- Citations: 0
- Source URL: https://arxiv.org/abs/2602.02425
- PDF: https://arxiv.org/pdf/2602.02425
- Provenance: arxiv

## Abstract

Protein fitness optimization is challenged by a vast combinatorial landscape where high-fitness variants are extremely sparse. Many current methods either underperform or require computationally expensive gradient-based sampling. We present CHASE, a framework that repurposes the evolutionary knowledge of pretrained protein language models by compressing their embeddings into a compact latent space. By training a conditional flow-matching model with classifier-free guidance, we enable the direct generation of high-fitness variants without predictor-based guidance during the ODE sampling steps. CHASE achieves state-of-the-art performance on AAV and GFP protein design benchmarks. Finally, we show that bootstrapping with synthetic data can further enhance performance in data-constrained settings.

_Abstract reproduced under CC0-1.0 (arXiv descriptive metadata)._

Topics: [AI Protein Design & Structure Prediction](https://space-kitty.com/topics/protein-design)

---
Space Kitty Archive · https://space-kitty.com · API: https://space-kitty.com/api/v1 · MCP: https://space-kitty.com/mcp · Premium: $0.20/request via x402 (Base Sepolia (testnet)). Metadata: arXiv (CC0), OpenAlex (CC0), Europe PMC (CC BY/CC0 works), ClinicalTrials.gov.