{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,31]],"date-time":"2026-08-31T09:18:36Z","timestamp":1788167916254,"version":"build-2803163510"},"reference-count":31,"publisher":"American Chemical Society (ACS)","issue":"8","license":[{"start":{"date-parts":[[2026,4,24]],"date-time":"2026-04-24T00:00:00Z","timestamp":1776988800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["22378328"],"award-info":[{"award-number":["22378328"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100012226","name":"Fundamental Research Funds for the Central Universities","doi-asserted-by":"publisher","award":["2025SMECP03"],"award-info":[{"award-number":["2025SMECP03"]}],"id":[{"id":"10.13039\/501100012226","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/na","name":"Young Science and Technology Rising Stars Project of Shaanxi Province","doi-asserted-by":"publisher","award":["2024ZC-KJXX-010"],"award-info":[{"award-number":["2024ZC-KJXX-010"]}],"id":[{"id":"10.13039\/na","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,8,24]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Formulation property prediction in industrial chemistry is often hindered by severe data scarcity due to high experimental costs and confidentiality constraints, which leads to overfitting and unreliable evaluation. Here, we propose a hybrid ResNet\u2013KAN framework for small-sample formulation modeling and demonstrate it on proprietary emulsion oil formulations (32 real formulations, 11 anonymized components) for predicting pH and defoaming performance. To improve data efficiency while preventing metric inflation, we integrate fold-wise CTGAN augmentation into a repeated nested cross-validation protocol, where synthetic (X, y) samples are generated exclusively within outer training folds and are never used for performance evaluation. Under this rigorous setting, the proposed approach achieves strong and stable generalization across both tasks, reaching R2 = 0.785 \u00b1 0.098 for pH and R2 = 0.755 \u00b1 0.068 for defoaming on outer test folds and consistently outperforming standard baselines and architectural variants. Beyond accuracy, interpretability is provided through complementary analyses: SHAP identifies globally influential components, while KAN-based functional and layer-level visualizations reveal explicit nonlinear response patterns and potential operating windows, enabling actionable formulation guidance without disclosing proprietary identities. Overall, this work presents a leakage-controlled augmentation and interpretable modeling strategy for data-scarce industrial formulations in precision chemistry.<\/jats:p>","DOI":"10.1021\/prechem.5c00402","type":"journal-article","created":{"date-parts":[[2026,4,24]],"date-time":"2026-04-24T09:51:42Z","timestamp":1777024302000},"page":"1190-1202","source":"Crossref","is-referenced-by-count":0,"title":["ResNet-KAN for Small-Sample Chemical Formulation Prediction:\nHybrid Architecture and Interpretability Analysis"],"prefix":"10.1021","volume":"4","author":[{"given":"Yaochen","family":"Zhang","sequence":"first","affiliation":[{"name":"Northwest University , , ,","place":["Xi\u2019an, China, 710069"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6230-0330","authenticated-orcid":true,"given":"Yuqi","family":"Wang","sequence":"additional","affiliation":[{"name":"Northwest University , , ,","place":["Xi\u2019an, China, 710069"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xin","family":"Ding","sequence":"additional","affiliation":[{"name":"Northwest University , , ,","place":["Xi\u2019an, China, 710069"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3361-0783","authenticated-orcid":true,"given":"Liang","family":"Zhao","sequence":"additional","affiliation":[{"name":"East China University of Science and Technology , , ,","place":["Shanghai, China, 200237"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0346-8376","authenticated-orcid":true,"given":"Le","family":"Wu","sequence":"additional","affiliation":[{"name":"Northwest University , , ,","place":["Xi\u2019an, China, 710069"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"316","published-online":{"date-parts":[[2026,4,24]]},"reference":[{"issue":"11","key":"2026083104362692200_cit1","doi-asserted-by":"publisher","first-page":"662","DOI":"10.1021\/prechem.5c00073","article-title":"Tumor Extracellular Vesicle Digital\nScoring Assays for Cancer Detection and Monitoring","volume":"3","author":"Lee","year":"2025","journal-title":"Precis. Chem."},{"issue":"11","key":"2026083104362692200_cit2","doi-asserted-by":"publisher","first-page":"570","DOI":"10.1021\/prechem.4c00051","article-title":"Leveraging Machine Learning Potentials for In-Situ\nSearching of Active Sites in Heterogeneous Catalysis","volume":"2","author":"Cheng","year":"2024","journal-title":"Precis. Chem."},{"key":"2026083104362692200_cit3","doi-asserted-by":"publisher","first-page":"206","DOI":"10.1021\/prechem.5c00088","article-title":"MODNet: A Monocular Object-Based\nDepth Estimation Network for AI Robotic Chemists","volume":"4","author":"Fu","year":"2025","journal-title":"Precis. Chem."},{"issue":"10","key":"2026083104362692200_cit4","doi-asserted-by":"publisher","first-page":"518","DOI":"10.1021\/prechem.4c00048","article-title":"Single-Molecule\nBioelectronic Sensors with AI-Aided\nData Analysis: Convergence and Challenges","volume":"2","author":"Yang","year":"2024","journal-title":"Precis.\nChem."},{"key":"2026083104362692200_cit5","doi-asserted-by":"publisher","first-page":"1003","DOI":"10.1016\/j.eng.2019.02.014","article-title":"Smart Process\nManufacturing for Formulated Products","volume":"5","author":"Litster","year":"2019","journal-title":"Engineering"},{"key":"2026083104362692200_cit6","doi-asserted-by":"publisher","first-page":"185","DOI":"10.1016\/j.compchemeng.2018.04.009","article-title":"Deep Convolutional\nNeural Network Model Based Chemical\nProcess Fault Diagnosis","volume":"115","author":"Wu","year":"2018","journal-title":"Comput. Chem. Eng."},{"key":"2026083104362692200_cit7","doi-asserted-by":"publisher","first-page":"2198","DOI":"10.1002\/aic.16157","article-title":"Machine Learning\nfor Crystal Identification and Discovery","volume":"64","author":"Spellings","year":"2018","journal-title":"AIChE\nJ."},{"key":"2026083104362692200_cit8","doi-asserted-by":"publisher","DOI":"10.1002\/aic.17041","article-title":"Machine Learning\nCorrected Alchemical Perturbation Density Functional Theory for Catalysis\nApplications","volume":"66","author":"Griego","year":"2020","journal-title":"AIChE J."},{"key":"2026083104362692200_cit9","doi-asserted-by":"publisher","first-page":"40","DOI":"10.1038\/s41580-021-00407-0","article-title":"A Guide to Machine\nLearning for Biologists","volume":"23","author":"Greener","year":"2022","journal-title":"Nat. Rev. Mol. Cell\nBiol."},{"key":"2026083104362692200_cit10","doi-asserted-by":"publisher","DOI":"10.1002\/aic.18333","article-title":"Data-Driven Identification of Crystallization\nKinetics","volume":"70","author":"Nyande","year":"2024","journal-title":"AIChE J."},{"key":"2026083104362692200_cit11","doi-asserted-by":"publisher","first-page":"8736","DOI":"10.1021\/acs.chemrev.3c00189","article-title":"Machine Learning Methods for Small Data Challenges\nin Molecular Science","volume":"123","author":"Dou","year":"2023","journal-title":"Chem. Rev."},{"key":"2026083104362692200_cit12","doi-asserted-by":"publisher","DOI":"10.1016\/j.ese.2022.100172","article-title":"Data Augmentation\nand Machine Learning Techniques for Control Strategy Development in\nBio-Polymerization Process","volume":"11","author":"Wei","year":"2022","journal-title":"Environ. Sci. Ecotechnol."},{"key":"2026083104362692200_cit13","doi-asserted-by":"publisher","DOI":"10.1016\/j.apenergy.2024.124844","article-title":"A Revolutionary\nNeural Network Architecture with Interpretability and Flexibility\nBased on Kolmogorov\u2013Arnold for Solar Radiation and Temperature\nForecasting","volume":"378","author":"Gao","year":"2025","journal-title":"Appl. Energy"},{"key":"2026083104362692200_cit14","doi-asserted-by":"publisher","DOI":"10.1002\/adem.202100612","article-title":"A Property-Driven\nStepwise Design\nStrategy for Multiple Low-Melting Alloys via Machine Learning","volume":"23","author":"Chen","year":"2021","journal-title":"Adv. Eng. Mater."},{"key":"2026083104362692200_cit15","doi-asserted-by":"publisher","first-page":"524","DOI":"10.1039\/D0CY01751E","article-title":"Direct Design of Active Catalysts\nfor Low Temperature Oxidative Coupling of Methane via Machine Learning\nand Data Mining","volume":"11","author":"Ohyama","year":"2021","journal-title":"Catal. Sci. Technol."},{"key":"2026083104362692200_cit16","doi-asserted-by":"publisher","DOI":"10.1016\/j.engappai.2025.110951","article-title":"Horizontal-to-Tilted Conversion of\nSolar Radiation Data Using Machine Learning Algorithms","volume":"153","author":"Celik","year":"2025","journal-title":"Engineering Applications of Artificial Intelligence"},{"key":"2026083104362692200_cit17","doi-asserted-by":"publisher","first-page":"383","DOI":"10.1007\/s41660-020-00151-6","article-title":"Resource\nAllocation in Multiple Energy-Integrated\nBiorefinery Using Neuroevolution and Mathematical Optimization","volume":"5","author":"Chan","year":"2021","journal-title":"Process Integr. Optim. Sustain."},{"key":"2026083104362692200_cit18","doi-asserted-by":"publisher","first-page":"599","DOI":"10.1007\/s41660-021-00163-w","article-title":"Comparison\nbetween Artificial Neural Networks and Support Vector Machine Modeling\nfor Polycaprolactone Synthesis via Enzyme-Catalyzed Polymerization","volume":"5","author":"Arumugasamy","year":"2021","journal-title":"Process Integr. Optim. Sustain."},{"key":"2026083104362692200_cit19","doi-asserted-by":"publisher","first-page":"728","DOI":"10.1038\/s41597-024-03573-w","article-title":"Accelerating Formulation\nDesign via\nMachine Learning: Generating a High-Throughput Shampoo Formulations\nDataset","volume":"11","author":"Chitre","year":"2024","journal-title":"Sci. Data"},{"key":"2026083104362692200_cit20","doi-asserted-by":"publisher","DOI":"10.1016\/j.aichem.2024.100054","article-title":"Machine Learning Assisted Analysis\nand Prediction of Rubber Formulation Using Existing Databases","volume":"2","author":"Deng","year":"2024","journal-title":"Artif. Intell. Chem."},{"key":"2026083104362692200_cit21","doi-asserted-by":"publisher","DOI":"10.1002\/aic.17990","article-title":"Machine\nLearning Enabled Integrated\nFormulation and Process Design Framework for a Pharmaceutical 3D Printing\nPlatform","volume":"69","author":"Sundarkumar","year":"2022","journal-title":"AIChE J."},{"key":"2026083104362692200_cit22","doi-asserted-by":"publisher","first-page":"126","DOI":"10.1002\/aic.690470113","article-title":"Artificial Neural-Network-Assisted\nStochastic Process Optimization Strategies","volume":"47","author":"Nandi","year":"2001","journal-title":"AIChE J."},{"key":"2026083104362692200_cit23","doi-asserted-by":"publisher","first-page":"42","DOI":"10.1038\/s41524-023-01000-z","article-title":"Small Data Machine Learning in Materials\nScience","volume":"9","author":"Xu","year":"2023","journal-title":"npj Comput. Mater."},{"key":"2026083104362692200_cit24","author":"Xu","year":"2019"},{"key":"2026083104362692200_cit25","doi-asserted-by":"publisher","first-page":"158820","DOI":"10.1109\/ACCESS.2019.2945545","article-title":"Comprehensive\nReview of Artificial\nNeural Network Applications to Pattern Recognition","volume":"7","author":"Abiodun","year":"2019","journal-title":"IEEE Access"},{"key":"2026083104362692200_cit26","doi-asserted-by":"publisher","DOI":"10.1016\/j.egyai.2025.100473","article-title":"Comparison of Kolmogorov\u2013Arnold\nNetworks and\nMulti-Layer Perceptron for Modelling and Optimisation Analysis of\nEnergy Systems","volume":"20","author":"Ansar","year":"2025","journal-title":"Energy AI"},{"key":"2026083104362692200_cit27","author":"Long","year":"2025"},{"key":"2026083104362692200_cit28","article-title":"A Comprehensive Survey on Kolmogorov\u2013Arnold\nNetworks","author":"Ji","year":"2025"},{"key":"2026083104362692200_cit29","author":"Ta","year":"2025"},{"key":"2026083104362692200_cit30","author":"Dorogush","year":"2017"},{"key":"2026083104362692200_cit31","unstructured":"Lundberg, S. M.; Lee, S.-I.\n          A Unified Approach\nto Interpreting Model Predictions.  Advances\nin Neural Information Processing Systems (NeurIPS 2017);\nLong Beach, CA, USA, December 4\u20139, 2017; vol 30; pp 4765\u20134774."}],"container-title":["Precision Chemistry"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/pubs.acs.org\/doi\/pdf\/10.1021\/prechem.5c00402","content-type":"application\/pdf","content-version":"vor","intended-application":"unspecified"},{"URL":"https:\/\/pubs.acs.org\/pcrhej\/article-pdf\/4\/8\/1190\/66264386\/prechem.5c00402.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/pubs.acs.org\/pcrhej\/article-pdf\/4\/8\/1190\/66264386\/prechem.5c00402.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,31]],"date-time":"2026-08-31T08:37:22Z","timestamp":1788165442000},"score":1,"resource":{"primary":{"URL":"https:\/\/pubs.acs.org\/pcrhej\/article\/4\/8\/1190\/5233040\/ResNet-KAN-for-Small-Sample-Chemical-Formulation"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,24]]},"references-count":31,"journal-issue":{"issue":"8","published-online":{"date-parts":[[2026,4,24]]},"published-print":{"date-parts":[[2026,8,24]]}},"URL":"https:\/\/doi.org\/10.1021\/prechem.5c00402","relation":{},"ISSN":["2771-9316"],"issn-type":[{"value":"2771-9316","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,24]]}}}