# Foundation-Model Supervision of Distilled Potentials for Million-Atom Molecular Dynamics

Preprint · Xiaoqing Liu, Yangshuai Wang, Lei Shen

- Archive id: arxiv:2610.23371
- Published: 2026-10-16
- arXiv: https://arxiv.org/abs/2610.23371
- Citations: 0
- Source URL: https://arxiv.org/abs/2610.23371
- PDF: https://arxiv.org/pdf/2610.23371
- Provenance: arxiv

## Abstract

Machine-learned interatomic potentials distilled from foundation models enable fast large-scale molecular dynamics (MD), but they can fail locally in chemical environments that their training data did not sample. System-averaged error metrics and small random audits often miss such sparse or isolated failures. Here we use an exact equivariant foundation model, MACE-MPA-0, to check every atom of million-atom production simulations. Reorganizing two-layer MACE inference into a receiver-chunked three-pass schedule that keeps no per-edge intermediates resident reduces the marginal memory per atom 340-fold, evaluates up to $2.86\\times10^{7}$ atoms on a single H200 GPU, and supervises $10^{6}$-atom trajectories in LAMMPS at 1.3% of the run time. Distilled MACE students run four to nine times slower than NEP89 with 0.95-1.28 times the teacher's force error against density functional theory (DFT). In molten SiO$\_2$, whole-system checks uncover a persistent tail of under-coordinated atoms hidden by the system-averaged error and an isolated, unphysical O-O contact that a 1% random audit misses 94% of the time. On-the-fly retraining on teacher-labelled 13-Å clusters, with the force loss restricted to their inner 6 Å, removes three quarters of the failure tail and the kilobar-level pressure biases in SiO$\_2$ and compressed water; the repair converges to the teacher, not to DFT.

_Abstract reproduced under CC0-1.0 (arXiv descriptive metadata)._

Topics: [AI for Science](https://space-kitty.com/topics/ai-for-science)

---
Space Kitty Archive · https://space-kitty.com · API: https://space-kitty.com/api/v1 · MCP: https://space-kitty.com/mcp · Premium: $0.20/request via x402 (Base Sepolia (testnet)). Metadata: arXiv (CC0), OpenAlex (CC0), Europe PMC (CC BY/CC0 works), ClinicalTrials.gov.