---
title: "APO：面向原子系统3D结构预测的无监督原子策略优化"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2607.28553v1"
permalink: "https://feed.astralor.com/items/itm_dee9fea3d377d91fc3200671"
published: "2026-07-30T17:21:58.000Z"
score: 77
channel: "ai"
tags: ["APO","FlowDPO","Flow Matching","无监督对齐","结构预测"]
---

# APO：面向原子系统3D结构预测的无监督原子策略优化

## AI 点评

> 这篇工作最值得关注的地方在于，它用物理一致性替代了昂贵的实验标签，相当于给无监督结构预测立了一个新标杆。对做材料或抗体计算的团队来说，APO直接解决数据匮乏的痛点；对做对齐算法的研究者，双奖励机制的设计也很有借鉴价值。

## 速览

预测原子系统的三维结构是材料科学与药物发现的基础。现有流匹配模型如FlowDPO依赖监督偏好学习对齐真实坐标，但实验标注成本高昂。本研究提出完全无监督的APO框架，无需真实参考结构，通过双奖励机制（特征增强与热力学稳定性）在采样组内实现自纠正，在晶体和抗体结构预测基准上超越全监督基线，并显著提升推理效率。

## 关键点

- APO完全无需标记坐标，通过双奖励机制实现无监督对齐。
- 在晶体与抗体结构预测基准上超越全监督方法，取得新SOTA。
- APO能拉直概率路径，显著提升推理效率。
- 验证了内在物理一致性比噪声监督坐标更适合作为对齐信号。

## 全文

Predicting the 3D structures of atomic systems is fundamental to advancing material science and drug discovery. While flow-matching models (, FlowDPO) have recently shown promise in this domain, their performance relies heavily on alignment with ground-truth coordinates via supervised preference learning. However, obtaining experimental labels for novel crystal phases or de novo proteins is prohibitively expensive, creating a bottleneck for structural modeling in data-scarce regimes. In this work, we propose (Atomic Policy Optimization), a fully unsupervised alignment framework that eliminates the need for ground-truth reference structures. APO adapts group-relative policy optimization to 3D atomic environments, utilizing a novel dual-reward mechanism: (i) a that reinforces the policy's dominant latent structural modes through eigen-decomposition of sample similarities, and (ii) a that enforces thermodynamic stability. Our framework enables the model to ``self-correct'' by identifying physically plausible configurations within sampled groups. Extensive benchmarks on crystal and antibody structure prediction demonstrate that APO consistently outperforms fully supervised baselines, achieving a new state-of-the-art in match rates and structural fidelity. Furthermore, we show that APO effectively straightens probability paths, significantly improving inference efficiency. Our results suggest that intrinsic physical consistency can serve as a superior guide for alignment compared to noisy, supervised coordinate matching.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_dee9fea3d377d91fc3200671