Research map / Architectures and efficiency / RNNs and normalization
MACE: Higher Order Equivariant Message Passing Neural Networks for Fast and Accurate Force Fields
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 changed | Where | Measured | Effect |
|---|---|---|---|
| Use higher correlation-order messages (ν=2 or ν=3) instead of ν=1 (ablation) | Two-layer invariant MACE on rMD17 aspirin · rMD17 aspirin learning curve | force MAE | Helped |
| Add equivariant messages with increasing symmetry order L at fixed ν=3 (ablation) | Two-layer MACE on rMD17 aspirin · rMD17 aspirin learning curve | force MAE learning-curve slope | No clear effect |
| Add equivariant messages with increasing symmetry order L at fixed ν=1 (ablation) | Two-layer MACE on rMD17 aspirin · rMD17 aspirin learning curve | force MAE learning-curve slope | No clear effect |
| Use higher-order messages rather than relying on additional two-body message-passing layers (ablation) | MACE versus two-body MPNNs on 3BPA · 3BPA | layers needed for accuracy convergence | Helped |
| Add nonlinearities in the update to increase body order (ablation) | NequIP versus BOTNet on 3BPA · 3BPA | accuracy | No clear effect |
| Increase equivariance from L=0 to L=2 (ablation) | MACE · 3BPA 1200 K | energy RMSE (meV): 37.1 (0.8) -> 29.8 (1.0) | Helped |
Closest papers by meaning
- ELoRA: Low-Rank Adaptation for Equivariant GNNs (ICML 2025 · similarity 0.88)
- Long-Short-Range Message-Passing: A Physics-Informed Framework to Capture Non-Local Interaction for Scalable Molecular Dynamics Simulation (ICLR 2024 · similarity 0.87)
- Higher-Rank Irreducible Cartesian Tensors for Equivariant Message Passing (NeurIPS 2024 · similarity 0.86)
- Compact SO(3) Equivariant Atomistic Foundation Models via Structural Pruning (NeurIPS 2026 · similarity 0.85)
- FlashTP: Fused, Sparsity-Aware Tensor Product for Machine Learning Interatomic Potentials (ICML 2025 · similarity 0.85)
- Towards Fast, Specialized Machine Learning Force Fields: Distilling Foundation Models via Energy Hessians (ICLR 2025 · similarity 0.84)
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.