Saved in:
Bibliographic Details
Main Authors: Wong, Nicolas, Yang, Julia H.
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.10159
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915851698962432
author Wong, Nicolas
Yang, Julia H.
author_facet Wong, Nicolas
Yang, Julia H.
contents Universal machine learned interatomic potentials (uMLIPs) embody a growing area of interest due to their transferability across the periodic table, displaying an error of about 0.6 kcal/mol against the Matbench Discovery test set. However, we show that achieving more accurate predictions on out-of-domain tasks requires fine-tuning. Additionally, we investigate the existence and influence of model biases in molecular dynamics (MD) by examining two approaches for data generation: from multiple MD trajectories in parallel, which we call naive fine-tuning, and from a single MD trajectory with fine-tuning after set intervals, which we call periodic fine-tuning. Our results find that naive fine-tuning generates constrained datasets that fail to represent MD simulations, and thus downstream fine-tuned models fail during extrapolation. In contrast, periodic fine-tuning yields models which are more generalizable and accurate, producing low-error dynamics. These findings indicate the role of uMLIP bias in fine-tuning, and highlights the need for multiple fine-tuning steps. Lastly, we relate unphysical behavior to principal component space, and quantify extrapolations through Q-residual analysis, which are useful as a proxy for epistemic uncertainty for larger simulations.
format Preprint
id arxiv_https___arxiv_org_abs_2603_10159
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Bias in Universal Machine-Learned Interatomic Potentials and its Effects on Fine-Tuning
Wong, Nicolas
Yang, Julia H.
Materials Science
Universal machine learned interatomic potentials (uMLIPs) embody a growing area of interest due to their transferability across the periodic table, displaying an error of about 0.6 kcal/mol against the Matbench Discovery test set. However, we show that achieving more accurate predictions on out-of-domain tasks requires fine-tuning. Additionally, we investigate the existence and influence of model biases in molecular dynamics (MD) by examining two approaches for data generation: from multiple MD trajectories in parallel, which we call naive fine-tuning, and from a single MD trajectory with fine-tuning after set intervals, which we call periodic fine-tuning. Our results find that naive fine-tuning generates constrained datasets that fail to represent MD simulations, and thus downstream fine-tuned models fail during extrapolation. In contrast, periodic fine-tuning yields models which are more generalizable and accurate, producing low-error dynamics. These findings indicate the role of uMLIP bias in fine-tuning, and highlights the need for multiple fine-tuning steps. Lastly, we relate unphysical behavior to principal component space, and quantify extrapolations through Q-residual analysis, which are useful as a proxy for epistemic uncertainty for larger simulations.
title Bias in Universal Machine-Learned Interatomic Potentials and its Effects on Fine-Tuning
topic Materials Science
url https://arxiv.org/abs/2603.10159