Title: Learning Inter-Atomic Potentials without Explicit Equivariance Show Chinese title

Title: 无需显式等变性的原子间势能学习

Abstract: Accurate and scalable machine-learned inter-atomic potentials (MLIPs) are essential for molecular simulations ranging from drug discovery to new material design. Current state-of-the-art models enforce roto-translational symmetries through equivariant neural network architectures, a hard-wired inductive bias that can often lead to reduced flexibility, computational efficiency, and scalability. In this work, we introduce TransIP: Transformer-based Inter-Atomic Potentials, a novel training paradigm for interatomic potentials achieving symmetry compliance without explicit architectural constraints. Our approach guides a generic non-equivariant Transformer-based model to learn SO(3)-equivariance by optimizing its representations in the embedding space. Trained on the recent Open Molecules (OMol25) collection, a large and diverse molecular dataset built specifically for MLIPs and covering different types of molecules (including small organics, biomolecular fragments, and electrolyte-like species), TransIP attains comparable performance in machine-learning force fields versus state-of-the-art equivariant baselines. Further, compared to a data augmentation baseline, TransIP achieves 40% to 60% improvement in performance across varying OMol25 dataset sizes. More broadly, our work shows that learned equivariance can be a powerful and efficient alternative to equivariant or augmentation-based MLIP models. Show Chinese abstract

Abstract: 准确且可扩展的机器学习原子间势能 (MLIPs) 对于从药物发现到新材料设计的分子模拟至关重要。当前最先进的模型通过等变神经网络架构强制旋转平移对称性，这是一种硬编码的归纳偏差，通常会导致灵活性、计算效率和可扩展性的降低。在本工作中，我们引入了 TransIP：基于 Transformer 的原子间势能，这是一种新的训练范式，用于实现对称性合规，而无需显式的架构约束。我们的方法引导一个通用的非等变基于 Transformer 的模型，在嵌入空间中优化其表示以学习 SO(3) 等变性。在最近的 Open Molecules (OMol25) 数据集上进行训练，这是一个为 MLIPs 专门构建的大而多样的分子数据集，涵盖了不同类型的分子（包括小有机物、生物分子片段和类似电解质的物质），TransIP 在机器学习力场方面的表现与最先进的等变基线相当。此外，与数据增强基线相比，TransIP 在不同大小的 OMol25 数据集上性能提高了 40% 到 60%。更广泛地说，我们的工作表明，学习的等变性可以成为等变或基于增强的 MLIP 模型的强大而高效的替代方案。