AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

Property-Driven Protein Inverse Folding with Multi-Objective Preference Alignment

Junqi Liu, Xiaoyang Hou, Chence Shi, Xin Liu, Zhi Yang, Jian Tang

ICLR 2026 · Probabilistic and scientific modeling › Molecules and proteins · approach: Structure-based 3D molecule generation · 3 citations (OpenAlex)

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 changedWhereMeasuredEffect
fine-tune ProteinMPNN with multi-objective semi-online DPO (MoMPNN Sol+TM)ProteinMPNN · CATH 4.3 testSolubility (Sol): 0.719 -> 0.884Improved
fine-tune ProteinMPNN with multi-objective DPO (MoMPNN Thermo+IG)ProteinMPNN · CATH 4.3 testThermostability (Thermo): 0.769 -> 0.963Improved
fine-tune with multi-objective DPO (Sol+TM)ProteinMPNN · CATH 4.3 testTM score: 0.740 -> 0.738No clear effect
use IG objective instead of TM for solubility alignment (ablation)MoMPNN · CATH 4.3 testRMSD: 4.37 -> 4.73Hurt
use IG instead of TM as designability objective (ablation)MoMPNN · De novo backbonesTM score: 0.734 -> 0.742 (Sol)Improved
add EP objective to Sol+TM (ablation)MoMPNN · CATH 4.3 testSolubility (Sol): 0.884 -> 0.852Hurt
add EP objective to Sol+TM (ablation)MoMPNN · CATH 4.3 testThermostability (Thermo): 0.747 -> 0.790Improved
MoMPNN versus ProteinMPNN on de novo backbonesProteinMPNN · De novo backbonesTM score: 0.718 -> 0.751Improved
train on property subset (HyperMPNN) rather than alignmentProteinMPNN · CATH 4.3 testTM score: 0.740 -> 0.706Hurt
multi-objective flexible-margin DPO instead of weighted-score DPO (ablation)MoMPNN · CATH 4.3 small validationEvo. ppl: 5.843 (weighted) vs 6.205 (MoMPNN)Mixed
multi-objective flexible-margin DPO instead of weighted-score DPO (ablation)MoMPNN · CATH 4.3 small validationTM score: 0.811 -> 0.825Improved
MoMPNN versus ProteinMPNN on binder designProteinMPNN · binder designsuccess rateImproved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy