NEAT模型让3D分子生成更高效且不依赖原子顺序。
NEAT: Neighborhood-Guided, Efficient, Autoregressive Set Transformer for 3D Molecular Generation
- 用邻域引导策略替代固定原子顺序，实现原子排列无关性。
- 在QM9和GEOM-Drugs数据集上达到当前最优生成质量。
- 生成速度显著快于现有方法，适合快速分子设计场景。
基于Transformer的自回归模型为生成3D分子提供了比扩散或流匹配方法更高效的替代方案。一个核心挑战是：标准Transformer需要对令牌进行序列排序，而分子中的原子并无固有顺序。以往工作通过使用规范原子顺序来解决此问题，但这些方法对原子排列不具有置换不变性，且使下一个原子的预测偏向于特定排序习惯。为此，本文提出新型邻域引导训练策略。所提出的NEAT（Neighborhood-Guided, Efficient, Autoregressive Set Transformer）将分子图视为原子集合，学习在图边界处的无序可接受令牌分布，从而实现原子级别的置换不变性。NEAT在QM9和GEOM-Drugs数据集上实现了当前最优的生成质量，并显著优于现有基线的速度表现。
原文摘要 · Abstract (English)
Transformer-based autoregressive models offer an efficient alternative to diffusion- and flow-matching-based approaches for generating 3D molecules. One challenge remains: standard transformer architectures require a sequential ordering of tokens, which is not inherently defined for the atoms in a molecule. Prior works have addressed this by using canonical atom orderings. However, these approaches are not permutation invariant w.r.t. atoms and bias next-token prediction towards ordering conventions. We overcome this limitation by introducing a novel neighborhood-guided training strategy. Our model, NEAT (Neighborhood-Guided, Efficient, Autoregressive Set Transformer) treats molecular graphs as sets of atoms and learns an order-agnostic distribution over admissible tokens at the graph boundary, thereby ensuring atom-level permutation invariance. NEAT achieves state-of-the-art generation quality on the QM9 and GEOM-Drugs datasets while offering a significant speed advantage over existing baselines.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。