mlsanityData to 14 Oct 2026

Property-driven Protein Inverse Folding With Multi-Objective Preference Alignment

arXiv 2603.06748 ↗First posted 6 Mar 2026cs.LG cs.AIcited by 3

Not on the map. Why?

Abstract (arXiv)

Protein sequence design must balance designability, defined as the ability to recover a target backbone, with multiple, often competing, developability properties such as solubility, thermostability, and expression. Existing approaches address these properties through post hoc mutation, inference-time biasing, or retraining on property-specific subsets, yet they are target dependent and demand substantial domain expertise or careful hyperparameter tuning. In this paper, we introduce ProtAlign, a multi-objective preference alignment framework that fine-tunes pretrained inverse folding models to satisfy diverse developability objectives while preserving structural fidelity. ProtAlign employs a semi-online Direct Preference Optimization strategy with a flexible preference margin to mitigate conflicts among competing objectives and constructs preference pairs using in silico property predictors. Applied to the widely used ProteinMPNN backbone, the resulting model MoMPNN enhances developability without compromising designability across tasks including sequence design for CATH 4.3 crystal structures, de novo generated backbones, and real-world binder design scenarios, making it an appealing framework for practical protein sequence design.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 31 of its 66 references that are in the map. First posted on arXiv 6 Mar 2026.

Builds on

31 of 66 references.

References · 31

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  3. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 8,065
  4. 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,720
  5. 2022Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference timecited by 1,396
  6. 2024SimPO: Simple Preference Optimization with a Reference-Free Rewardcited by 963
  7. 2023A General Theoretical Paradigm to Understand Learning from Human Preferencescited by 912
  8. 2023Fine-Grained Human Feedback Gives Better Rewards for Language Model Trainingcited by 401
  9. 2020Learning from Protein Structure with Geometric Vector Perceptronscited by 313
  10. 2025LLaDA 1.5: Variance-Reduced Preference Optimization for Large Language Diffusion Modelscited by 285
  11. 2023Personalized Soups: Personalized Large Language Model Alignment via Post-hoc Parameter Mergingcited by 280
  12. 2024Direct Language Model Alignment from Online AI Feedbackcited by 260
19 more
  1. 2023Rewarded soups: towards Pareto-optimal alignment by interpolating weights fine-tuned on diverse rewardscited by 243
  2. 2023Mitigating the Alignment Tax of RLHFcited by 183
  3. 2024Rewards-in-Context: Multi-objective Alignment of Foundation Models with Dynamic Preference Adjustmentcited by 175
  4. 2024Unlocking Guidance for Discrete State-Space Diffusion and Flow Modelscited by 129
  5. 2024Understanding the performance gap between online and offline alignment algorithmscited by 110
  6. 2024Controllable Preference Optimization: Toward Controllable Multi-Objective Alignmentcited by 107
  7. 2024Human Alignment of Large Language Models through Online Preference Optimisationcited by 91
  8. 2022PiFold: Toward effective and efficient protein inverse foldingcited by 83
  9. 2023Structure-informed Language Models Are Protein Designerscited by 68
  10. 2022PD-MORL: Preference-Driven Multi-Objective Reinforcement Learning Algorithmcited by 58
  11. 2024Conditional Language Policy: A General Framework for Steerable Multi-Objective Finetuningcited by 43
  12. 2023Scaling Pareto-Efficient Decision Making Via Offline Multi-Objective RLcited by 36
  13. 2025Bridging Offline and Online Reinforcement Learning for LLMscited by 25
  14. 2025Robust Multi-Objective Preference Alignment with Online DPOcited by 22
  15. 2024SurfPro: Functional Protein Design Based on Continuous Surfacecited by 14
  16. 2023Free from Bellman Completeness: Trajectory Stitching via Model-based Return-conditioned Supervised Learningcited by 13
  17. 2025Improving Protein Sequence Design through Designability Preference Optimizationcited by 6
  18. 2025ProteinZero: Self-Improving Protein Generation via Online Reinforcement Learningcited by 6
  19. 2025Protein Inverse Folding From Structure Feedbackcited by 5

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 3.

3 papers.

  1. 2026Generative Modeling in Protein Design: Neural Representations, Conditional Generation, and Evaluation Standardscited by 1
  2. 2026ProteinOPD: Towards Effective and Efficient Preference Alignment for Protein Designcited by 0
  3. 2026ZetaDial: dialing net charge of protein binders at inference time for therapeutic developabilitycited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Improving Inverse Folding for Peptide Design with Diversity-regularized Direct Preference Optimization
  2. 2025ProtInvTree: Deliberate Protein Inverse Folding with Reward-guided Tree Search
  3. 2026RosettaSearch: Multi-Objective Inference-Time Search for Protein Sequence Design
  4. 2025EnerBridge-DPO: Energy-Guided Protein Inverse Folding with Markov Bridges and Direct Preference Optimization
  5. 2024Reinforcement learning on structure-conditioned categorical diffusion for protein inverse folding
  6. 2022AlphaFold Distillation for Protein Design
  7. 2025PRISM: Enhancing Protein Inverse Folding through Fine-Grained Retrieval on Structure-Sequence Multimodal Representations
  8. 2024ProteinWeaver: A Divide-and-Assembly Approach for Protein Backbone Design