AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

EPO: Diverse and Realistic Protein Ensemble Generation via Energy Preference Optimization

Yuancheng Sun, Yuxuan Ren, Zhaoming Chen, Xu Han, Kang Liu, Qiwei Ye

AAAI 2026 · Probabilistic and scientific modeling › Molecules and proteins · approach: Generative protein structure design · 2 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
use listwise instead of pairwise energy preference loss (ablation)EPO · Fast-Folding Rg JSD (mean/median)JSD: 2.68/2.54 -> 2.36/2.28Improved
use listwise instead of pairwise energy preference loss (ablation)EPO · Fast-Folding RMSF JSDJSD: 0.39/0.40 -> 0.32/0.30Improved
use listwise instead of pairwise energy preference loss (ablation)EPO · TetrapeptidesJSD TICA-0,1: 0.318 -> 0.311Improved
use listwise instead of pairwise energy preference loss (ablation)EPO · ATLAS%PC-sim>0.5: 9 -> 13Improved
use SDE instead of ODE sampling during online optimization (ablation)EPO · TetrapeptidesJSD chi_all: 0.132 -> 0.107Improved
use SDE instead of ODE sampling during online optimization (ablation)EPO · Tetrapeptides TICA-0,1JSD: 0.337 -> 0.311Improved
apply EPO online energy preference fine-tuning to pretrained MDGenMDGen · ATLASExposed residue J: 0.29 -> 0.44Improved
apply EPO fine-tuning to pretrained MDGenMDGen · ATLASWeak contacts J: 0.51 -> 0.45Hurt
apply EPO fine-tuning to pretrained MDGenMDGen · ATLASRoot W2: 2.69/3.47 -> 2.79/3.43Hurt
apply EPO fine-tuning to pretrained MDGenMDGen · TetrapeptidesJSD chi_all: 0.109 -> 0.107Improved
use a very small beta (ablation)EPO · Tetrapeptidesdistribution qualityHurt
use too few denoising steps in SDE sampling (ablation)EPO · Tetrapeptidestraining stabilityHurt

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