# PFArena: Benchmarking Language Models for Protein Modification

arXiv: https://arxiv.org/abs/2609.28921 · HTML: https://mlsanity.com/paper/2609.28921/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 8 of its 40 references that are in the map.
- First posted on arXiv 24 Sep 2026.

## Builds on (most cited first, 8 of 8)

- [DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence](https://mlsanity.com/paper/2606.19348/) (2026)
- [ProGen2: Exploring the Boundaries of Protein Language Models](https://mlsanity.com/paper/2206.13517/) (2022)
- [Tranception: protein fitness prediction with autoregressive transformers and inference-time retrieval](https://mlsanity.com/paper/2205.13760/) (2022)
- [ProtAgents: Protein discovery via large language model multi-agent collaborations combining physics and machine learning](https://mlsanity.com/paper/2402.04268/) (2024)
- [Multi-Scale Representation Learning for Protein Fitness Prediction](https://mlsanity.com/paper/2412.01108/) (2024)
- [Large Language Model is Secretly a Protein Sequence Optimizer](https://mlsanity.com/paper/2501.09274/) (2025)
- [Towards A Generative Protein Evolution Machine with DPLM-Evo](https://mlsanity.com/paper/2605.00182/) (2026)
- [AMix-2: Establishing Protein as a Native Modality in Large Language Models](https://mlsanity.com/paper/2605.30963/) (2026)

## Abstract (arXiv)

Protein modification requires navigating an immense sequence space, yet wet-lab validation remains low-throughput and costly. Although computational paradigms including protein language models (PLMs), large language models (LLMs), and LLM-based agents have shown promise in protein modification, their relative efficacy across realistic experimental decision-making settings remains unclear. To bridge this gap, we introduce PFArena, a benchmark comprising four controlled task interfaces that cover single-mutant generation and multi-mutant ranking. By providing varying levels of mutation fitness data, PFArena reflects four representative research scenarios characterized by differing degrees of prior experimental context. We assess six PLMs, six LLMs, and five LLM-based agents using complementary metrics to measure both peak and overall protein modification performance. Our evaluation reveals that model performance shifts systematically with the availability of target-specific experimental evidence: PLMs demonstrate proficiency in open-ended single-mutant generation by leveraging protein-specific priors, whereas LLMs and agents perform strongly in multi-mutant ranking, particularly when target-specific fitness data are available. Nevertheless, all model families face fundamental challenges with increasing search-space size and mutation depth. We release our code and benchmark suite to facilitate reproducible research in model-assisted protein modification.