Skip to content
DraconiticResearch
Search Sources API
197,768 studies31 sourcesUpdated 24 min ago
PreprintCC BYChemRxiv

A review on foundation models for polymer informatics

  • Yihan Zhu1,2
  • Eric Inae1,2
  • Yuhan Liu3,2
  • Weijiang Li1,2
  • Sobin Alosious4,2,5
  • Renzheng Zhang4,2
  • Gang Liu1,2
  • Tengfei Luo4,2,5,6
  • Meng Jiang1,2,5
6 affiliations
  1. 1Department of Computer Science and Engineering
  2. 2University of Notre Dame
  3. 3Department of Chemistry and Biochemistry
  4. 4Department of Aerospace and Mechanical Engineering
  5. 5Lucy Family Institute for Data and Society
  6. 6Department of Chemical and Biomolecular Engineering
Open at sourcePDFDOICite

Abstract

Foundation models have emerged as a powerful paradigm for learning transferable knowledge from broad, heterogeneous data and adapting them to diverse downstream tasks, driving a growing interest in AI-accelerated scientific discovery. In polymer science, this paradigm offers a promising route beyond conventional machine learning methods that are often confined to isolated tasks and struggle to generalize under data scarcity and across heterogeneous downstream settings. Polymer foundation models (PFMs), however, are not merely general scientific foundation models applied to polymer datasets. They must address representation, generation, transfer, and evaluation challenges specific to polymer informatics. This review synthesizes recent progress in PFMs for representation learning and generative modeling, spanning models designed for polymer systems and those adapted from general foundation models. We organize this review around two stages: (1) polymer modeling as the foundation, covering representations, datasets, model architectures, and pretraining objectives; and (2) downstream applications, evaluation, and adaptation. By synthesizing recent progress, open challenges, and future directions, this review outlines a roadmap toward robust and generalizable foundation models for polymer informatics.

Related

  1. Polymer Informatics in the Era of Foundation Models: From Representation to Autonomous Design
    Jianxin He, Ying Li
    ChemRxivPreprintCC BY-NC-NDFull textMaterials & polymersChemRxiv
  2. AdaptDelivery: a polymer informatics platform for predicting structural properties of polymers via machine learning
    Lorand Gabriel Parajdi, Istvan Toth, Alex-Adrian Farcas +3
    Scientific ReportsPeer-reviewedCC BY 4.0Full textMaterials & polymersCrossref
  3. Polymer Informatics Atlas: Property-Specific Representations, Chemistry-Aware Generalization, and Reliability-Aware Machine Learning Across Five Polymer Properties
    Dennis Obinna Orji
    ChemRxivPreprintCC BYFull textMaterials & polymersChemRxiv
  4. PolyNet: A Python Package for Polymer Informatics
    Eduardo Aguilar-Bejarano, Morgan R Alexander, David Winkler +1
    ChemRxivPreprintCC BYMaterials & polymersChemRxiv
  5. Polymer-based heterojunction photocatalysts for environmental remediation and solar fuels: a review on interface engineering, stability, and sustainable design
    Abdul Majid, Muhammad Tayyab Raza, Anzal Noureen +2
    RSC advancesPeer-reviewedCC BYFull textGeneral chemistryEurope PMC

Cite

BibTeXRIS
BibTeX
@misc{zhu2026review,
  title = {{A review on foundation models for polymer informatics}},
  author = {Yihan Zhu and Eric Inae and Yuhan Liu and Weijiang Li and Sobin Alosious and Renzheng Zhang and Gang Liu and Tengfei Luo and Meng Jiang},
  howpublished = {ChemRxiv},
  year = {2026},
  month = oct,
  doi = {10.26434/chemrxiv.15009949/v1},
  url = {https://chemrxiv.org/doi/full/10.26434/chemrxiv.15009949/v1},
  note = {CC BY},
}
RIS
TY  - UNPB
TI  - A review on foundation models for polymer informatics
AU  - Yihan Zhu
AU  - Eric Inae
AU  - Yuhan Liu
AU  - Weijiang Li
AU  - Sobin Alosious
AU  - Renzheng Zhang
AU  - Gang Liu
AU  - Tengfei Luo
AU  - Meng Jiang
PY  - 2026
DA  - 2026/10/05
PB  - ChemRxiv
DO  - 10.26434/chemrxiv.15009949/v1
UR  - https://chemrxiv.org/doi/full/10.26434/chemrxiv.15009949/v1
AB  - Foundation models have emerged as a powerful paradigm for learning transferable knowledge from broad, heterogeneous data and adapting them to diverse downstream tasks, driving a growing interest in AI-accelerated scientific discovery. In polymer science, this paradigm offers a promising route beyond conventional machine learning methods that are often confined to isolated tasks and struggle to generalize under data scarcity and across heterogeneous downstream settings. Polymer foundation models (PFMs), however, are not merely general scientific foundation models applied to polymer datasets. They must address representation, generation, transfer, and evaluation challenges specific to polymer informatics. This review synthesizes recent progress in PFMs for representation learning and generative modeling, spanning models designed for polymer systems and those adapted from general foundation models. We organize this review around two stages: (1) polymer modeling as the foundation, covering representations, datasets, model architectures, and pretraining objectives; and (2) downstream applications, evaluation, and adaptation. By synthesizing recent progress, open challenges, and future directions, this review outlines a roadmap toward robust and generalizable foundation models for polymer informatics.
ER  -