{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,14]],"date-time":"2026-10-14T11:49:21Z","timestamp":1791978561755,"version":"4.4.0"},"reference-count":54,"publisher":"AIP Publishing","issue":"14","funder":[{"DOI":"10.13039\/501100014896","name":"Graduate School, National University of Defense Technology","doi-asserted-by":"publisher","award":["No. 25-XXXX-XX-CXXM-01"],"award-info":[{"award-number":["No. 25-XXXX-XX-CXXM-01"]}],"id":[{"id":"10.13039\/501100014896","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["pubs.aip.org"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2026,10,14]]},"abstract":"<jats:p>Graph neural networks have become a central tool for polymer property prediction, but their fixed activation functions can impose a low-pass spectral bias that smooths over the sharp, multi-scale structure\u2013property variations that distinguish chemically similar polymers. We address this limitation with PolyFKAN, a multitask graph-learning architecture that augments the prediction head of a polyGNN backbone with Fourier Kolmogorov\u2013Arnold network layers. PolyFKAN represents edge-wise activations as truncated Fourier series and couples the resulting spectral branch to the standard multilayer perceptron (MLP) pathway through a bounded residual gate, thereby increasing functional expressivity without disrupting the learned graph embeddings. Across 17 polymer properties spanning electronic, thermodynamic, optical\/dielectric, thermal, and gas-permeability domains, PolyFKAN achieves statistically significant improvement over a tightly controlled polyGNN baseline on 11 of 17 targets (one-sided t-test, p &amp;lt; 0.05), with one additional target showing marginal improvement. The gains are especially pronounced for multi-scale properties, including more than 15% average root mean squared error reduction for large-penetrant gas permeabilities and more than 15% improvement across optical and dielectric targets. Hierarchical ablation further reveals that optimal Fourier bandwidth is property-dependent: difficult, multi-scale targets require broader spectral capacity with stronger amplitude control, whereas high-fidelity targets benefit from lighter regularization. These results establish Fourier-KAN prediction heads as a minimally invasive and physically aligned inductive bias for polymer informatics.<\/jats:p>","DOI":"10.1063\/5.0346557","type":"journal-article","created":{"date-parts":[[2026,10,14]],"date-time":"2026-10-14T11:01:13Z","timestamp":1791975673000},"update-policy":"https:\/\/doi.org\/10.1063\/aip-crossmark-policy-page","source":"Crossref","is-referenced-by-count":0,"title":["PolyFKAN: Integrating Fourier-KAN layers into graph neural networks for polymer property prediction"],"prefix":"10.1063","volume":"165","author":[{"ORCID":"https:\/\/orcid.org\/0009-0008-5967-8509","authenticated-orcid":false,"given":"Boran","family":"Li","sequence":"first","affiliation":[{"name":"College of Science, National University of Defense Technology , Changsha,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0337-588X","authenticated-orcid":false,"given":"Jiying","family":"Liu","sequence":"additional","affiliation":[{"name":"College of Science, National University of Defense Technology , Changsha,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5401-3494","authenticated-orcid":false,"given":"Qi","family":"Yu","sequence":"additional","affiliation":[{"name":"College of Science, National University of Defense Technology , Changsha,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-6997-2125","authenticated-orcid":false,"given":"Zexi","family":"Yang","sequence":"additional","affiliation":[{"name":"College of Science, National University of Defense Technology , Changsha,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-1857-1643","authenticated-orcid":false,"given":"Yapeng","family":"Zhan","sequence":"additional","affiliation":[{"name":"College of Science, National University of Defense Technology , Changsha,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"317","published-online":{"date-parts":[[2026,10,14]]},"reference":[{"key":"2026101407010775200_c1","doi-asserted-by":"publisher","first-page":"236","DOI":"10.1016\/j.pmatsci.2016.06.002","article-title":"Advanced polymeric dielectrics for high energy density applications","volume":"83","year":"2016","journal-title":"Prog. Mater. Sci."},{"key":"2026101407010775200_c2","doi-asserted-by":"publisher","first-page":"511","DOI":"10.1038\/nmat2745","volume":"9","year":"2010","journal-title":"Nat. Mater."},{"key":"2026101407010775200_c3","doi-asserted-by":"publisher","first-page":"23","DOI":"10.1016\/s0376-7388(96)00061-0","volume":"125","year":"1997","journal-title":"J.\u00a0Membr. Sci."},{"key":"2026101407010775200_c4","doi-asserted-by":"publisher","first-page":"4845","DOI":"10.1038\/ncomms5845","volume":"5","year":"2014","journal-title":"Nat. Commun."},{"key":"2026101407010775200_c5","doi-asserted-by":"publisher","first-page":"37","DOI":"10.1002\/masy.19991470106","volume":"147","year":"1999","journal-title":"Macromol. Symp."},{"key":"2026101407010775200_c6","doi-asserted-by":"publisher","first-page":"2075","DOI":"10.1039\/c2cs35350d","volume":"42","year":"2013","journal-title":"Chem. Soc. Rev."},{"key":"2026101407010775200_c7","doi-asserted-by":"publisher","first-page":"160012","DOI":"10.1038\/sdata.2016.12","volume":"3","year":"2016","journal-title":"Sci. Data"},{"key":"2026101407010775200_c8","doi-asserted-by":"publisher","first-page":"1078","DOI":"10.1021\/acsmacrolett.7b00228","volume":"6","year":"2017","journal-title":"ACS Macro Lett."},{"key":"2026101407010775200_c9","doi-asserted-by":"publisher","first-page":"100595","DOI":"10.1016\/j.mser.2020.100595","volume":"144","year":"2021","journal-title":"Mater. Sci. Eng., R"},{"key":"2026101407010775200_c10","doi-asserted-by":"publisher","first-page":"655","DOI":"10.1038\/s41578-020-00255-y","volume":"6","year":"2021","journal-title":"Nat. Rev. Mater."},{"key":"2026101407010775200_c11","doi-asserted-by":"publisher","first-page":"17575","DOI":"10.1021\/acs.jpcc.8b02913","volume":"122","year":"2018","journal-title":"J.\u00a0Phys. Chem. C"},{"key":"2026101407010775200_c12","doi-asserted-by":"publisher","first-page":"171104","DOI":"10.1063\/5.0023759","volume":"128","year":"2020","journal-title":"J.\u00a0Appl. Phys."},{"key":"2026101407010775200_c13","doi-asserted-by":"publisher","first-page":"044902","DOI":"10.1063\/5.0012283","volume":"128","year":"2020","journal-title":"J.\u00a0Appl. Phys."},{"key":"2026101407010775200_c14","doi-asserted-by":"publisher","first-page":"742","DOI":"10.1021\/ci100050t","volume":"50","year":"2010","journal-title":"J.\u00a0Chem. Inf. Model."},{"key":"2026101407010775200_c15","unstructured":"Y.\n            Yang\n          , \u201cEnhancing polymer informatics: A study on the impact of fingerprinting methods and machine learning models,\u201d Ph.D. thesis, Cornell University, 2023."},{"key":"2026101407010775200_c16","doi-asserted-by":"publisher","first-page":"5838","DOI":"10.1016\/j.matpr.2021.02.730","article-title":"Deep learning based approach for prediction of glass transition temperature in polymers","volume":"46","year":"2021","journal-title":"Mater. Today: Proc."},{"key":"2026101407010775200_c17","doi-asserted-by":"publisher","first-page":"64","DOI":"10.1038\/s41524-023-01016-5","volume":"9","year":"2023","journal-title":"npj Comput. Mater."},{"key":"2026101407010775200_c18","doi-asserted-by":"publisher","first-page":"4099","DOI":"10.1038\/s41467-023-39868-6","volume":"14","year":"2023","journal-title":"Nat. Commun."},{"key":"2026101407010775200_c19","doi-asserted-by":"publisher","first-page":"7002","DOI":"10.1021\/acs.chemmater.5c00940","volume":"37","year":"2025","journal-title":"Chem. Mater."},{"key":"2026101407010775200_c20","doi-asserted-by":"publisher","first-page":"234111","DOI":"10.1063\/1.5099132","volume":"150","year":"2019","journal-title":"J.\u00a0Chem. Phys."},{"key":"2026101407010775200_c21","doi-asserted-by":"publisher","first-page":"10486","DOI":"10.1039\/d2sc02839e","volume":"13","year":"2022","journal-title":"Chem. Sci."},{"key":"2026101407010775200_c22","doi-asserted-by":"publisher","first-page":"1560","DOI":"10.1021\/acs.chemmater.2c02991","volume":"35","year":"2023","journal-title":"Chem. Mater."},{"key":"2026101407010775200_c23","first-page":"5301","year":"2019"},{"key":"2026101407010775200_c24","first-page":"685","year":"2020"},{"key":"2026101407010775200_c25","first-page":"7537","volume-title":"Advances in Neural Information Processing Systems (NeurIPS)","year":"2020"},{"key":"2026101407010775200_c26","article-title":"A unified theory of sinusoidal activation families for implicit neural representations","year":"2026"},{"key":"2026101407010775200_c27","unstructured":"J.\n            Sahs\n          , R.Pyle, F.Anselmi, and A. B.Patel, arXiv:2503.10587 (2025)."},{"key":"2026101407010775200_c28","first-page":"66342","article-title":"KAN: Kolmogorov-Arnold Networks","year":"2025"},{"key":"2026101407010775200_c29","first-page":"953","volume":"114","year":"1957","journal-title":"Dokl. Akad. Nauk SSSR"},{"key":"2026101407010775200_c30","doi-asserted-by":"publisher","first-page":"119","DOI":"10.1016\/j.neunet.2021.01.020","volume":"137","year":"2021","journal-title":"Neural Networks"},{"key":"2026101407010775200_c31","unstructured":"Z.\n            Bozorgasl\n           and H.Chen, arXiv:2405.12832 (2024)."},{"key":"2026101407010775200_c32","doi-asserted-by":"publisher","first-page":"130525","DOI":"10.1109\/ACCESS.2025.3566551","volume":"13","year":"2025","journal-title":"IEEE Access"},{"key":"2026101407010775200_c33","doi-asserted-by":"publisher","first-page":"104319","DOI":"10.1016\/j.cag.2025.104319","article-title":"PointNet with KAN versus PointNet with MLP for 3D classification and segmentation of point sets","volume":"131","year":"2025","journal-title":"Computers & Graphics"},{"key":"2026101407010775200_c34","unstructured":"R.\n            Genet\n           and H.Inzirillo, arXiv:2405.07344 (2024)."},{"key":"2026101407010775200_c35","unstructured":"M.\n            Cheon\n          , arXiv:2406.00600 (2024)."},{"key":"2026101407010775200_c36","doi-asserted-by":"publisher","first-page":"134313","DOI":"10.1016\/j.jhydrol.2025.134313","article-title":"A graph Fourier Kolmogorov-Arnold Network (G-FourierKAN) and its application to spring discharge simulation","volume":"664","year":"2026","journal-title":"J.\u00a0Hydrol."},{"key":"2026101407010775200_c37","unstructured":"M.\n            Kiamari\n          , M.Kiamari, and B.Krishnamachari, arXiv:2406.06470 (2024)."},{"key":"2026101407010775200_c38","unstructured":"N.\n            Volzhin\n           and S.Yoon, arXiv:2512.19494 (2025)."},{"key":"2026101407010775200_c39","doi-asserted-by":"publisher","first-page":"1346","DOI":"10.1038\/s42256-025-01087-7","volume":"7","year":"2025","journal-title":"Nat. Mach. Intell."},{"key":"2026101407010775200_c40","doi-asserted-by":"publisher","first-page":"5435","DOI":"10.1021\/acs.jcim.2c00875","volume":"62","year":"2022","journal-title":"J.\u00a0Chem. Inf. Model."},{"key":"2026101407010775200_c41","first-page":"112","volume":"95","year":"2001","journal-title":"Chem. Listy"},{"key":"2026101407010775200_c42","doi-asserted-by":"publisher","first-page":"31","DOI":"10.1021\/ci00057a005","volume":"28","year":"1988","journal-title":"J.\u00a0Chem. Inf. Comput. Sci."},{"key":"2026101407010775200_c43","doi-asserted-by":"publisher","first-page":"3515","DOI":"10.1021\/acs.macromol.4c00508","volume":"57","year":"2024","journal-title":"Macromolecules"},{"key":"2026101407010775200_c44","article-title":"PolyidTM (polymer inverse design)","volume-title":"NREL Software Record, SWR-21-98","year":"2022"},{"key":"2026101407010775200_c45","first-page":"1263","year":"2017"},{"key":"2026101407010775200_c46"},{"key":"2026101407010775200_c47","first-page":"3438","year":"2020"},{"key":"2026101407010775200_c48","first-page":"3538","year":"2018"},{"key":"2026101407010775200_c49","first-page":"xiii","volume-title":"Chebyshev Polynomials: From Approximation Theory to Algebra and Number Theory","year":"1990"},{"key":"2026101407010775200_c50","doi-asserted-by":"publisher","first-page":"135","DOI":"10.1007\/bf02392815","volume":"116","year":"1966","journal-title":"Acta Math."},{"key":"2026101407010775200_c51","doi-asserted-by":"publisher","first-page":"744","DOI":"10.1090\/s0002-9904-1971-12793-3","volume":"77","year":"1971","journal-title":"Bull. Am. Math. Soc."},{"key":"2026101407010775200_c52","doi-asserted-by":"publisher","first-page":"513","DOI":"10.1039\/c7sc02664a","volume":"9","year":"2018","journal-title":"Chem. Sci."},{"key":"2026101407010775200_c53","doi-asserted-by":"publisher","first-page":"109","DOI":"10.1080\/14686996.2019.1567402","volume":"20","year":"2019","journal-title":"Sci. Technol. Adv. Mater."},{"key":"2026101407010775200_c54","doi-asserted-by":"publisher","first-page":"186","DOI":"10.1038\/s41524-024-01373-9","volume":"10","year":"2024","journal-title":"npj Comput. Mater."}],"container-title":["The Journal of Chemical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/pubs.aip.org\/aip\/jcp\/article-pdf\/doi\/10.1063\/5.0346557\/21170376\/144905_1_5.0346557.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/pubs.aip.org\/aip\/jcp\/article-pdf\/doi\/10.1063\/5.0346557\/21170376\/144905_1_5.0346557.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,14]],"date-time":"2026-10-14T11:01:19Z","timestamp":1791975679000},"score":1,"resource":{"primary":{"URL":"https:\/\/pubs.aip.org\/jcp\/article\/165\/14\/144905\/3406503\/PolyFKAN-Integrating-Fourier-KAN-layers-into-graph"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,10,14]]},"references-count":54,"journal-issue":{"issue":"14","published-print":{"date-parts":[[2026,10,14]]}},"URL":"https:\/\/doi.org\/10.1063\/5.0346557","relation":{},"ISSN":["0021-9606","1089-7690"],"issn-type":[{"value":"0021-9606","type":"print"},{"value":"1089-7690","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026,10,14]]},"published":{"date-parts":[[2026,10,14]]},"article-number":"144905"}}