# Preference-based Antibody Expression Ranking: Scaling with Large-scale Weak Supervision

arXiv: https://arxiv.org/abs/2607.16263 · HTML: https://mlsanity.com/paper/2607.16263/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Preference optimization for molecular design, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 8 of its 35 references that are in the map.
- First posted on arXiv 29 Jun 2026.

## Placed in

- LLM alignment / RLHF / preference optimisation / [Preference optimization for molecular design](https://mlsanity.com/alignment/preference-optimization-for-molecular-design/)

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

- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Antigen-Specific Antibody Design via Direct Energy-based Preference Optimization](https://mlsanity.com/paper/2403.16576/) (2024)
- [Large scale paired antibody language models](https://mlsanity.com/paper/2403.17889/) (2024)
- [Pareto-Optimal Energy Alignment for Designing Nature-Like Antibodies](https://mlsanity.com/paper/2412.20984/) (2024)
- [Improving Protein Sequence Design through Designability Preference Optimization](https://mlsanity.com/paper/2506.00297/) (2025)
- [Protein Inverse Folding From Structure Feedback](https://mlsanity.com/paper/2506.03028/) (2025)
- [A SARS-CoV-2 Interaction Dataset and VHH Sequence Corpus for Antibody Language Models](https://mlsanity.com/paper/2405.18749/) (2024)
- [Structure-Aware Antibody Design with Affinity-Optimized Inverse Folding](https://mlsanity.com/paper/2512.17815/) (2025)

## Abstract (arXiv)

Antibody expression ranking is a critical task in antibody design, yet its modelling is severely hindered by the scarcity of labeled experimental data. To address this, we propose a unified preference-based learning framework that integrates scarce quantitative expression data with large-scale weak positive supervision from immunization data. We adapt Direct Preference Optimization (DPO) to protein language models by introducing a union-masked log-likelihood approximation and IMGT-based alignment, enabling efficient training on variable-length sequences. Evaluating on a diverse internal dataset of 1254 labeled sequences and 4 million unlabeled camelid-derived antibodies, we show that our method consistently outperforms baselines on most metrics. Our results demonstrate that preference learning can effectively learn from weak supervision, providing a scalable solution for antibody expressibility optimization in data-constrained settings. Project page: https://kisoji-biotechnology-inc.github.io/Preference-Expression-Ranking/.