Research map / Large language models / Alignment and preferences
Preference-based Antibody Expression Ranking: Scaling with Large-scale Weak Supervision
Josh Qixuan Sun, Morteza Babaie, Wenyang hou, Mark Crowley, David Young
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add Stage I continual pretraining before preference optimization (ablation) | Ours (AntiBERTa2) · Exp-1K test MCC | MCC: 26.7 -> 43.7 | Improved |
| add Stage I continual pretraining before preference optimization (ablation) | Ours (AntiBERTa2) · Exp-1K test Tau | Kendall Tau: 33.8 -> 46.3 | Improved |
| add Stage I continual pretraining before supervised regression (ablation) | Supervised (AntiBERTa2) · Exp-1K test composite Avg | Avg: 33.1 -> 29.0 | Hurt |
| add Stage I continual pretraining before supervised regression (ablation) | Supervised (IgBERT) · Exp-1K test composite Avg | Avg: 17.2 -> 37.5 | Improved |
| replace supervised regression with preference optimization on same Stage I backbone | AntiBERTa2 · Exp-1K test composite Avg | Avg: 29.0 -> 45.0 | Improved |
| add Camel-4M weak supervision to DPO (ablation) | Ours (AntiBERTa2) · Exp-1K MCC | MCC: 32.8 -> 43.7 | Improved |
| add Camel-4M weak supervision to DPO (ablation) | Ours (AntiBERTa2) · Exp-1K Tau | Kendall Tau: 35.1 -> 46.3 | Improved |
| use DPO instead of regression on labeled data only (no Camel-4M) | AntiBERTa2 · Exp-1K Tau | Kendall Tau: 37.6 -> 35.1 | Hurt |
| raise sampling rate alpha from 10 to 1000 (ablation) | Ours · Exp-1K Tau | Kendall Tau: 12.2 -> 40.5 | Improved |
| raise sampling rate alpha from 1000 to 10000 (ablation) | Ours · Exp-1K Tau | Kendall Tau: 40.5 -> 30.7 | Hurt |
| filter preference pairs by similarity threshold 0.59 (ablation) | Ours (AntiBERTa2) · Exp-1K MCC | MCC: 43.7 -> 6.4 | Hurt |
| add weak data to reduce generalization gap (ablation) | Ours · |Avg val - Avg test| | delta Avg: 14.1 -> 5.5 | Improved |
Closest papers by meaning
- Pareto-Optimal Energy Alignment for Designing Nature-Like Antibodies (NeurIPS 2025 · similarity 0.90)
- Antigen-Specific Antibody Design via Direct Energy-based Preference Optimization (NeurIPS 2024 · similarity 0.88)
- Bayesian Optimization of Antibodies Informed by a Generative Model of Evolving Sequences (ICLR 2025 · similarity 0.86)
- Antibody Design Using a Score-based Diffusion Model Guided by Evolutionary, Physical and Geometric Constraints (ICML 2024 · similarity 0.86)
- Retrieval Augmented Diffusion Model for Structure-informed Antibody Design and Optimization (ICLR 2025 · similarity 0.86)
- AsEP: Benchmarking Deep Learning Methods for Antibody-specific Epitope Prediction (NeurIPS 2024 · similarity 0.86)
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.
