aiqbench analysis
Zatom-2 represents a significant advancement in atomistic generative modeling by leveraging a multiscale Transformer architecture to unify data across chemistry, materials science, and biology. This approach effectively bridges the gap between data-rich organic domains and data-scarce biological contexts, demonstrating improved fidelity in molecular distribution and enhanced protein generation capabilities through transfer learning. The model's ability to handle diverse force regimes and scale from small molecules to complex proteins marks a crucial step toward general-purpose tools for accelerating scientific discovery.
Zatom-2: Multitask Pretraining on Atomistic Data for Generative Modeling across Domains
arXiv:2610.11454v1 Announce Type: cross Abstract: Unified atomistic modeling has the potential to accelerate discovery in chemistry, materials science, and biology by bridging data-rich chemical domains and data-scarce biological contexts. However, existing generative approaches to atomistic modeling remain highly specialized to scientific disciplines (chemistry vs. biology) or do not leverage both high-volume organic (molecule) and inorganic (material) data for general-purpose pretraining. To this end, we introduce Zatom-2, an atomistic generative model pretrained on approximately five million structures from the OMol25 and OMat24 electronic structure datasets. Zatom-2 features a multiscale Transformer architecture coupled with conditional flow matching that supports force conditioning and foundational pretraining tasks such as generation, structure prediction, and prediction of molecular and material energies and forces. Empirically, Zatom-2 achieves better molecular distribution fidelity than Zatom-1 and achieves strong performance on existing molecule and material generation benchmarks. Zatom-2 demonstrates the ability to control sample generation across low- and high-force regimes, and enhances protein generation in a low-data setting through joint generative-predictive pretraining and transfer learning, increasing protein backbone designability in a length extrapolation setting from 67.8% without pretraining to 74.8% after finetuning on 2,000 protein domains.
Authors:Miruna Cretu, Alex Abrudan, Antonia Panescu, Tynan Perez, Rishabh Anand, N. Benjamin Erichson, Michael W. Mahoney, Samuel Blau, Joseph Jacobson, Rafael Gómez-Bombarelli, Rex Ying, Tuomas Knowles, Pietro Liò, Alex Morehead
Abstract:Unified atomistic modeling has the potential to accelerate discovery in chemistry, materials science, and biology by bridging data-rich chemical domains and data-scarce biological contexts. However, existing generative approaches to atomistic modeling remain highly specialized to scientific disciplines (chemistry vs. biology) or do not leverage both high-volume organic (molecule) and inorganic (material) data for general-purpose pretraining. To this end, we introduce Zatom-2, an atomistic generative model pretrained on approximately five million structures from the OMol25 and OMat24 electronic structure datasets. Zatom-2 features a multiscale Transformer architecture coupled with conditional flow matching that supports force conditioning and foundational pretraining tasks such as generation, structure prediction, and prediction of molecular and material energies and forces. Empirically, Zatom-2 achieves better molecular distribution fidelity than Zatom-1 and achieves strong performance on existing molecule and material generation benchmarks. Zatom-2 demonstrates the ability to control sample generation across low- and high-force regimes, and enhances protein generation in a low-data setting through joint generative-predictive pretraining and transfer learning, increasing protein backbone designability in a length extrapolation setting from 67.8% without pretraining to 74.8% after finetuning on 2,000 protein domains.
Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
arXiv:2610.11454 [cs.LG]
(or arXiv:2610.11454v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2610.11454 arXiv-issued DOI via DataCite (pending registration)
From: N. Benjamin Erichson [view email] [v1] Thu, 8 Oct 2026 08:09:17 UTC (11,304 KB)
Full text under the publisher's open licence · 247 words
Related coverage
LeRobot v0.6.0: Imagine, Evaluate, Improve
Hugging Face · Jul 7, 2026
A Unified Vision-Language Model for PSMA PET/CT Report Generation, Visual Question Answering, and Lesion Segmentation
arXiv · Sep 16, 2026
How AI Coding Agents Can Unlock Materials Simulation with NVIDIA ALCHEMI Toolkit
NVIDIA · Aug 18, 2026
Develop Humanoid Robot Policies End-to-End with NVIDIA Isaac GR00T
NVIDIA · Jul 7, 2026