{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,22]],"date-time":"2026-10-22T15:43:14Z","timestamp":1792683794663,"version":"4.5.0"},"reference-count":72,"publisher":"American Chemical Society (ACS)","issue":"7","license":[{"start":{"date-parts":[[2023,3,23]],"date-time":"2023-03-23T00:00:00Z","timestamp":1679529600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100013348","name":"Innosuisse - Schweizerische Agentur f?r Innovationsf?rderung","doi-asserted-by":"publisher","award":["46966.1"],"award-info":[{"award-number":["46966.1"]}],"id":[{"id":"10.13039\/501100013348","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2023,4,10]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>The prediction of chemical reaction pathways has been accelerated by the development of novel machine learning architectures based on the deep learning paradigm. In this context, deep neural networks initially designed for language translation have been used to accurately predict a wide range of chemical reactions. Among models suited for the task of language translation, the recently introduced molecular transformer reached impressive performance in terms of forward-synthesis and retrosynthesis predictions. In this study, we first present an analysis of the performance of transformer models for product, reactant, and reagent prediction tasks under different scenarios of data availability and data augmentation. We find that the impact of data augmentation depends on the prediction task and on the metric used to evaluate the model performance. Second, we probe the contribution of different combinations of input formats, tokenization schemes, and embedding strategies to model performance. We find that less stable input settings generally lead to better performance. Lastly, we validate the superiority of round-trip accuracy over simpler evaluation metrics, such as top-k accuracy, using a committee of human experts and show a strong agreement for predictions that pass the round-trip test. This demonstrates the usefulness of more elaborate metrics in complex predictive scenarios and highlights the limitations of direct comparisons to a predefined database, which may include a limited number of chemical reaction pathways.<\/jats:p>","DOI":"10.1021\/acs.jcim.2c01407","type":"journal-article","created":{"date-parts":[[2023,3,23]],"date-time":"2023-03-23T14:33:18Z","timestamp":1679581998000},"page":"1914-1924","source":"Crossref","is-referenced-by-count":28,"title":["Transformer Performance\nfor Chemical Reactions: Analysis\nof Different Predictive and Evaluation Scenarios"],"prefix":"10.1021","volume":"63","author":[{"given":"Fernando","family":"Jaume-Santero","sequence":"first","affiliation":[{"name":"University of Geneva , , ,","place":["Geneva, Switzerland, 1205"]},{"name":"HES-SO University of Applied Sciences and Arts of Western Switzerland , , ,","place":["Geneva, Switzerland, 1227"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7266-627X","authenticated-orcid":true,"given":"Alban","family":"Bornet","sequence":"additional","affiliation":[{"name":"University of Geneva , , ,","place":["Geneva, Switzerland, 1205"]},{"name":"HES-SO University of Applied Sciences and Arts of Western Switzerland , , ,","place":["Geneva, Switzerland, 1227"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Alain","family":"Valery","sequence":"additional","affiliation":[{"name":"SpiroChem AG , ,","place":["Basel, Switzerland, 4058"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nona","family":"Naderi","sequence":"additional","affiliation":[{"name":"HES-SO University of Applied Sciences and Arts of Western Switzerland , , ,","place":["Geneva, Switzerland, 1227"]},{"name":"Swiss Institute of Bioinformatics , ,","place":["Lausanne, Switzerland, 1015"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"David","family":"Vicente Alvarez","sequence":"additional","affiliation":[{"name":"University of Geneva , , ,","place":["Geneva, Switzerland, 1205"]},{"name":"HES-SO University of Applied Sciences and Arts of Western Switzerland , , ,","place":["Geneva, Switzerland, 1227"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1089-8726","authenticated-orcid":true,"given":"Dimitrios","family":"Proios","sequence":"additional","affiliation":[{"name":"University of Geneva , , ,","place":["Geneva, Switzerland, 1205"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3309-6128","authenticated-orcid":true,"given":"Anthony","family":"Yazdani","sequence":"additional","affiliation":[{"name":"University of Geneva , , ,","place":["Geneva, Switzerland, 1205"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5568-4110","authenticated-orcid":true,"given":"Colin","family":"Bournez","sequence":"additional","affiliation":[{"name":"SpiroChem AG , ,","place":["Basel, Switzerland, 4058"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Thomas","family":"Fessard","sequence":"additional","affiliation":[{"name":"SpiroChem AG , ,","place":["Basel, Switzerland, 4058"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Douglas","family":"Teodoro","sequence":"additional","affiliation":[{"name":"University of Geneva , , ,","place":["Geneva, Switzerland, 1205"]},{"name":"HES-SO University of Applied Sciences and Arts of Western Switzerland , , ,","place":["Geneva, Switzerland, 1227"]},{"name":"Swiss Institute of Bioinformatics , ,","place":["Lausanne, Switzerland, 1015"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"316","published-online":{"date-parts":[[2023,3,23]]},"reference":[{"key":"2026081804434806900_cit1","doi-asserted-by":"publisher","first-page":"533","DOI":"10.1038\/nrd.2016.29","article-title":"Small Molecules, Big\nTargets: Drug Discovery Faces the Protein\u2013Protein Interaction\nChallenge","volume":"15","author":"Scott","year":"2016","journal-title":"Nat. Rev. Drug Discovery"},{"key":"2026081804434806900_cit2","doi-asserted-by":"publisher","first-page":"1560","DOI":"10.1016\/j.drudis.2022.02.015","article-title":"Small Molecules\nand Their Impact in Drug Discovery: A Perspective on the Occasion\nof the 125th Anniversary of the Bayer Chemical Research Laboratory","volume":"27","author":"Beck","year":"2022","journal-title":"Drug Discovery Today"},{"key":"2026081804434806900_cit3","doi-asserted-by":"publisher","first-page":"374","DOI":"10.1021\/ci0255782","article-title":"Cheminformatics\nAnalysis of Organic Substituents: Identification of the Most Common\nSubstituents, Calculation of Substituent Properties, and Automatic\nIdentification of Drug-like Bioisosteric Groups","volume":"43","author":"Ertl","year":"2003","journal-title":"J. Chem. Inf. Comput. Sci."},{"key":"2026081804434806900_cit4","doi-asserted-by":"publisher","first-page":"3","DOI":"10.1002\/(sici)1098-1128(199601)16:1&lt;3::aid-med1&gt;3.0.co;2-6","article-title":"The Art\nand Practice of Structure-based\nDrug Design: A Molecular Modeling Perspective","volume":"16","author":"Bohacek","year":"1996","journal-title":"Med. Res. Rev."},{"key":"2026081804434806900_cit5","doi-asserted-by":"publisher","first-page":"675","DOI":"10.1007\/s10822-013-9672-4","article-title":"Estimation of the Size of Drug-like\nChemical Space Based on GDB-17 Data","volume":"27","author":"Polishchuk","year":"2013","journal-title":"J. Comput.-Aided\nMol. Des."},{"key":"2026081804434806900_cit6","doi-asserted-by":"publisher","first-page":"225","DOI":"10.1021\/acsphyschemau.1c00051","article-title":"New Tools\nfor Taming Complex Reaction Networks: The Unimolecular Decomposition\nof Indole Revisited","volume":"2","author":"Garay-Ruiz","year":"2022","journal-title":"ACS Phys. Chem. Au"},{"key":"2026081804434806900_cit7","doi-asserted-by":"publisher","first-page":"10520","DOI":"10.1021\/acs.chemrev.8b00728","article-title":"Concepts of Artificial Intelligence\nfor Computer-Assisted Drug Discovery","volume":"119","author":"Yang","year":"2019","journal-title":"Chem.\nRev."},{"key":"2026081804434806900_cit8","doi-asserted-by":"publisher","first-page":"10001","DOI":"10.1021\/acs.chemrev.0c01303","article-title":"Ab Initio\nMachine Learning in Chemical Compound Space","volume":"121","author":"Huang","year":"2021","journal-title":"Chem. Rev."},{"key":"2026081804434806900_cit9","doi-asserted-by":"publisher","first-page":"547","DOI":"10.1038\/s41586-018-0337-2","article-title":"Machine Learning\nfor Molecular and Materials Science","volume":"559","author":"Butler","year":"2018","journal-title":"Nature"},{"key":"2026081804434806900_cit10","doi-asserted-by":"publisher","first-page":"604","DOI":"10.1038\/nature25978","article-title":"Planning\nChemical Syntheses with\nDeep Neural Networks and Symbolic AI","volume":"555","author":"Segler","year":"2018","journal-title":"Nature"},{"key":"2026081804434806900_cit11","doi-asserted-by":"publisher","first-page":"1465","DOI":"10.1021\/acscentsci.8b00357","article-title":"Using Machine Learning To Predict Suitable Conditions\nfor Organic Reactions","volume":"4","author":"Gao","year":"2018","journal-title":"ACS Cent. Sci."},{"key":"2026081804434806900_cit12","doi-asserted-by":"publisher","first-page":"153","DOI":"10.1007\/978-1-0716-1787-8_6","article-title":"Has Artificial Intelligence Impacted\nDrug Discovery?","volume":"2390","author":"Patronov","year":"2022","journal-title":"Methods Mol. Biol."},{"key":"2026081804434806900_cit13","doi-asserted-by":"publisher","first-page":"360","DOI":"10.1126\/science.aat2663","article-title":"Inverse Molecular\nDesign Using Machine Learning: Generative\nModels for Matter Engineering","volume":"361","author":"Sanchez-Lengeling","year":"2018","journal-title":"Science"},{"key":"2026081804434806900_cit14","doi-asserted-by":"publisher","first-page":"3156","DOI":"10.1038\/s41467-021-23415-2","article-title":"Masked Graph\nModeling\nfor Molecule Generation","volume":"12","author":"Mahmood","year":"2021","journal-title":"Nat. Commun."},{"key":"2026081804434806900_cit15","doi-asserted-by":"publisher","first-page":"015016","DOI":"10.1088\/2632-2153\/abc81d","article-title":"Prediction\nof Chemical\nReaction Yields Using Deep Learning","volume":"2","author":"Schwaller","year":"2021","journal-title":"Mach. learn.:\nsci. technol."},{"key":"2026081804434806900_cit16","doi-asserted-by":"publisher","first-page":"144","DOI":"10.1038\/s42256-020-00284-w","article-title":"Mapping the Space\nof Chemical Reactions\nUsing Attention-Based Neural Networks","volume":"3","author":"Schwaller","year":"2021","journal-title":"Nat.\nMach. Intell."},{"key":"2026081804434806900_cit17","doi-asserted-by":"publisher","first-page":"91","DOI":"10.1039\/D1DD00006C","article-title":"Reaction Classification and Yield\nPrediction Using the Differential Reaction Fingerprint DRFP","volume":"1","author":"Probst","year":"2022","journal-title":"Digit. Discovery"},{"key":"2026081804434806900_cit18","doi-asserted-by":"publisher","first-page":"1572","DOI":"10.1021\/acscentsci.9b00576","article-title":"Molecular\nTransformer: A Model for\nUncertainty-Calibrated Chemical Reaction Prediction","volume":"5","author":"Schwaller","year":"2019","journal-title":"ACS Cent. Sci."},{"key":"2026081804434806900_cit19","doi-asserted-by":"publisher","first-page":"370","DOI":"10.1039\/C8SC04228D","article-title":"A Graph-Convolutional Neural Network\nModel for the Prediction of Chemical Reactivity","volume":"10","author":"Coley","year":"2019","journal-title":"Chem. Sci."},{"key":"2026081804434806900_cit20","doi-asserted-by":"publisher","first-page":"685","DOI":"10.1246\/bcsj.20200017","article-title":"Quantum Chemical\nReaction Prediction\nMethod Based on Machine Learning","volume":"93","author":"Fujinami","year":"2020","journal-title":"Bull. Chem.\nSoc. Jpn."},{"key":"2026081804434806900_cit21","doi-asserted-by":"publisher","first-page":"5966","DOI":"10.1002\/chem.201605499","article-title":"Neural-Symbolic\nMachine Learning for Retrosynthesis\nand Reaction Prediction","volume":"23","author":"Segler","year":"2017","journal-title":"Chem.\ue5f8Eur. J."},{"key":"2026081804434806900_cit22","doi-asserted-by":"publisher","first-page":"12152","DOI":"10.1039\/C9CC05122H","article-title":"Molecular Transformer\nUnifies Reaction\nPrediction and Retrosynthesis across Pharma Chemical Space","volume":"55","author":"Lee","year":"2019","journal-title":"Chem. Commun."},{"key":"2026081804434806900_cit23","doi-asserted-by":"publisher","first-page":"3316","DOI":"10.1039\/C9SC05704H","article-title":"Predicting\nRetrosynthetic Pathways Using Transformer-Based Models and a Hyper-Graph\nExploration Strategy","volume":"11","author":"Schwaller","year":"2020","journal-title":"Chem. Sci."},{"key":"2026081804434806900_cit24","doi-asserted-by":"publisher","first-page":"5575","DOI":"10.1038\/s41467-020-19266-y","article-title":"State-of-the-Art Augmented\nNLP Transformer Models for Direct and Single-Step Retrosynthesis","volume":"11","author":"Tetko","year":"2020","journal-title":"Nat. Commun."},{"key":"2026081804434806900_cit25","doi-asserted-by":"publisher","first-page":"1612","DOI":"10.1021\/jacsau.1c00246","article-title":"Deep Retrosynthetic\nReaction Prediction Using Local Reactivity and\nGlobal Attention","volume":"1","author":"Chen","year":"2021","journal-title":"JACS Au"},{"key":"2026081804434806900_cit26","doi-asserted-by":"publisher","first-page":"55","DOI":"10.1186\/s13321-020-00458-z","article-title":"EvoMol: A\nFlexible and Interpretable Evolutionary Algorithm for Unbiased de\nNovo Molecular Generation","volume":"12","author":"Leguy","year":"2020","journal-title":"J. Cheminf."},{"key":"2026081804434806900_cit27","doi-asserted-by":"publisher","first-page":"186","DOI":"10.1126\/science.aar5169","article-title":"Predicting\nReaction Performance in C\u2013N Cross-Coupling Using Machine Learning","volume":"360","author":"Ahneman","year":"2018","journal-title":"Science"},{"key":"2026081804434806900_cit28","doi-asserted-by":"publisher","DOI":"10.1126\/sciadv.abe4166","article-title":"Extraction\nof Organic Chemistry Grammar from Unsupervised Learning of Chemical\nReactions","volume":"7","author":"Schwaller","year":"2021","journal-title":"Sci. Adv."},{"key":"2026081804434806900_cit29","doi-asserted-by":"publisher","first-page":"bbab340","DOI":"10.1093\/bib\/bbab340","article-title":"Graph Representation\nLearning in Bioinformatics: Trends, Methods and Applications","volume":"23","author":"Yi","year":"2022","journal-title":"Briefings Bioinf."},{"key":"2026081804434806900_cit30","doi-asserted-by":"publisher","first-page":"31","DOI":"10.1021\/ci00057a005","article-title":"SMILES, a\nChemical Language and Information System. 1. Introduction to Methodology\nand Encoding Rules","volume":"28","author":"Weininger","year":"1988","journal-title":"J. Chem. Inf. Model."},{"key":"2026081804434806900_cit31","doi-asserted-by":"publisher","first-page":"045024","DOI":"10.1088\/2632-2153\/aba947","article-title":"Self-Referencing Embedded Strings (SELFIES): A 100%\nRobust Molecular String Representation","volume":"1","author":"Krenn","year":"2020","journal-title":"Mach.\nlearn.: sci. technol."},{"key":"2026081804434806900_cit32","doi-asserted-by":"publisher","DOI":"10.26434\/chemrxiv.7097960.v1","article-title":"DeepSMILES: An Adaptation\nof SMILES for Use in Machine-Learning of\nChemical Structures","author":"O\u2019Boyle","year":"2018","journal-title":"ChemRxiv"},{"key":"2026081804434806900_cit33","doi-asserted-by":"publisher","first-page":"22","DOI":"10.1186\/1758-2946-4-22","article-title":"Towards\na Universal SMILES Representation-A Standard Method to Generate Canonical\nSMILES Based on the InChI","volume":"4","author":"O\u2019Boyle","year":"2012","journal-title":"J. Cheminf."},{"key":"2026081804434806900_cit34","doi-asserted-by":"publisher","first-page":"7","DOI":"10.1186\/1758-2946-5-7","article-title":"InChI-the Worldwide Chemical Structure Identifier Standard","volume":"5","author":"Heller","year":"2013","journal-title":"J. Cheminf."},{"key":"2026081804434806900_cit35","doi-asserted-by":"crossref","DOI":"10.1016\/j.patter.2022.100588","article-title":"SELFIES and\nthe Future of Molecular\nString Representations","author":"Krenn","year":"2022"},{"key":"2026081804434806900_cit36","doi-asserted-by":"crossref","unstructured":"Jawahar, G.; Sagot, B.; Seddah, D.\n          What Does BERT Learn about\nthe Structure of Language?. Proceedings of\nthe 57th Annual Meeting of the Association for Computational Linguistics; Association for Computational Linguistics: Stroudsburg, PA, USA, 2019; pp 3651\u20133657.","DOI":"10.18653\/v1\/P19-1356"},{"key":"2026081804434806900_cit37","doi-asserted-by":"crossref","unstructured":"Yenicelik, D.; Schmidt, F.; Kilcher, Y.\n          How Does BERT\nCapture Semantics? A Closer Look at Polysemous Words. Proceedings of the Third BlackboxNLP Workshop on Analyzing\nand Interpreting Neural Networks for NLP; Association for Computational Linguistics: Stroudsburg, PA, USA, 2020; pp 156\u2013162.","DOI":"10.18653\/v1\/2020.blackboxnlp-1.15"},{"key":"2026081804434806900_cit38","unstructured":"Lowe, D. M.\n          \n          Extraction\nof Chemical Structures and Reactions from the Literature, Ph.D. Thesis, University of Cambridge, 2012."},{"key":"2026081804434806900_cit39","doi-asserted-by":"crossref","unstructured":"He, J.; Quoc Nguyen, D.; Akhondi, S. A.; Druckenbrodt, C.; Thorne, C.; Hoessel, R.; Afzal, Z.; Zhai, Z.; Fang, B.; Yoshikawa, H.; Albahem, A.; Wang, J.; Ren, Y.; Zhang, Z.; Zhang, Y.; Hoang Dao, M.; Ruas, P.; Lamurias, A.; Couto, F. M.; Copara Zea, J. L.; Naderi, N.; Knafou, J. D. M.; Ruch, P.; Teodoro, D.; Lowe, D. M.; Mayfield, J.; Ko\u0308ksal, A.; Do\u0308nmez, H.; O\u0308zkirimli, E.; O\u0308zgu\u0308r, A.; Mahendran, D.; Gurdin, G.; Lewinski, N.; Tang, C.; McInnes, B. T.; Malarkodi, C. S.; Rk Rao, P.; Lalitha Devi, S.; Cavedon, L.; Cohn, T.; Baldwin, T.; Verspoor, K.\n          An Extended Overview of the CLEF 2020\nChEMU Lab: Information Extraction of Chemical Reactions from Patents. Proceedings of CLEF (Conference and Labs of the Evaluation\nForum) 2020 Working Notes; CEUR Workshop Proceedings (CEUR-WS.org): Thessaloniki, Greece, 2020.","DOI":"10.1007\/978-3-030-58219-7_18"},{"key":"2026081804434806900_cit40","unstructured":"Lowe, D. M.\n          \n          Chemical Reactions\nfrom US Patents (1976-Sep2016). Figshare Dataset, 2017.10.6084\/m9.figshare.5104873.v1."},{"key":"2026081804434806900_cit41","doi-asserted-by":"publisher","first-page":"100219","DOI":"10.1016\/j.ceja.2021.100219","article-title":"Reaction Engineering with Recurrent\nNeural Network: Kinetic Study\nof Dushman Reaction","volume":"9","author":"Murakami","year":"2022","journal-title":"Chem. Eng. J. Adv."},{"key":"2026081804434806900_cit42","doi-asserted-by":"publisher","first-page":"3178","DOI":"10.1038\/s41598-021-81889-y","article-title":"Discovery of Novel\nChemical Reactions\nby Deep Generative Recurrent Neural Network","volume":"11","author":"Bort","year":"2021","journal-title":"Sci. Rep."},{"key":"2026081804434806900_cit43","unstructured":"Vaswani, A.; Shazeer, N.; Parmar, N.; Uszkoreit, J.; Jones, L.; Gomez, A. N.; Kaiser, \u0141.; Polosukhin, I.\n          Attention\nIs All You Need. 31st Conference on Neural\nInformation Processing Systems\n(NIPS 2017), 2017; pp 5998\u20136008."},{"key":"2026081804434806900_cit44","article-title":"A Survey of Transformers","author":"Lin","year":"2021"},{"key":"2026081804434806900_cit45","doi-asserted-by":"publisher","first-page":"1371","DOI":"10.1039\/c9ra08535a","article-title":"Retrosynthesis\nwith attention-based NMT model and chemical analysis of \u201cwrong\u201d\npredictions","volume":"10","author":"Duan","year":"2020","journal-title":"RSC Adv."},{"key":"2026081804434806900_cit46","doi-asserted-by":"crossref","unstructured":"Karpov, P.; Godin, G.; Tetko, I. V.\n          A Transformer Model for Retrosynthesis. International Conference on Artificial Neural Networks; Springer, 2019; pp 817\u2013830.","DOI":"10.1007\/978-3-030-30493-5_78"},{"key":"2026081804434806900_cit47","doi-asserted-by":"publisher","first-page":"47","DOI":"10.1021\/acs.jcim.9b00949","article-title":"Predicting Retrosynthetic\nReactions Using Self-Corrected Transformer Neural Networks","volume":"60","author":"Zheng","year":"2019","journal-title":"J. Chem. Inf. Model."},{"key":"2026081804434806900_cit48","doi-asserted-by":"publisher","first-page":"2357","DOI":"10.3390\/molecules25102357","article-title":"Transfer Learning: Making Retrosynthetic\nPredictions Based on a Small\nChemical Reaction Dataset Scale to a New Level","volume":"25","author":"Bai","year":"2020","journal-title":"Molecules"},{"key":"2026081804434806900_cit49","doi-asserted-by":"crossref","DOI":"10.26434\/chemrxiv.8168354.v1","article-title":"Automatic Retrosynthetic Pathway\nPlanning Using Template-Free\nModels","author":"Lin","year":"2019"},{"key":"2026081804434806900_cit50","doi-asserted-by":"crossref","DOI":"10.26434\/chemrxiv.9897365.v2","article-title":"Data-Driven\nChemical Reaction Classification, Fingerprinting and Clustering Using\nAttention-Based Neural Networks","author":"Schwaller","year":"2019"},{"key":"2026081804434806900_cit51","doi-asserted-by":"publisher","first-page":"3398","DOI":"10.1021\/acs.jcim.0c00403","article-title":"Data Augmentation\nand Pretraining for Template-Based Retrosynthetic Prediction in Computer-Aided\nSynthesis Planning","volume":"60","author":"Fortunato","year":"2020","journal-title":"J. Chem. Inf. Model."},{"key":"2026081804434806900_cit52","article-title":"Neural Machine\nTranslation\nof Rare Words with Subword Units","author":"Sennrich","year":"2015"},{"key":"2026081804434806900_cit53","doi-asserted-by":"crossref","unstructured":"Schuster, M.; Nakajima, K.\n          Japanese and Korean\nVoice Search. 2012 IEEE international conference\non acoustics, speech and signal processing (ICASSP); IEEE, 2012; pp 5149\u20135152.","DOI":"10.1109\/ICASSP.2012.6289079"},{"key":"2026081804434806900_cit54","doi-asserted-by":"crossref","DOI":"10.18653\/v1\/D18-2012","article-title":"Sentencepiece:\nA Simple and Language Independent Subword Tokenizer and Detokenizer\nfor Neural Text Processing","author":"Kudo","year":"2018"},{"key":"2026081804434806900_cit55","unstructured":"Summary of the tokenizers. https:\/\/huggingface.co\/docs\/transformers\/tokenizer_summary (accessed\nOct 12, 2022)."},{"key":"2026081804434806900_cit56","doi-asserted-by":"publisher","first-page":"1560","DOI":"10.1021\/acs.jcim.0c01127","article-title":"SMILES Pair Encoding:\nA Data-Driven Substructure Tokenization\nAlgorithm for Deep Learning","volume":"61","author":"Li","year":"2021","journal-title":"J. Chem. Inf. Model."},{"key":"2026081804434806900_cit57","unstructured":"Socher, R.; Bauer, J.; Manning, C. D.; Ng, A.\nY.\n          Parsing with Compositional\nVector\nGrammars. Proceedings of the 51st Annual Meeting\nof the Association for Computational Linguistics; Long Papers, 2013; Vol. 1, pp 455\u2013465."},{"key":"2026081804434806900_cit58","doi-asserted-by":"crossref","unstructured":"Socher, R.; Perelygin, A.; Wu, J.; Chuang, J.; Manning, C. D.; Ng, A. Y.; Potts, C.\n          Recursive Deep Models for Semantic Compositionality\nover a Sentiment Treebank. Proceedings of\nthe 2013 conference on empirical methods in natural language processing, 2013; pp 1631\u20131642.","DOI":"10.18653\/v1\/D13-1170"},{"key":"2026081804434806900_cit59","unstructured":"Mikolov, T.; Chen, K.; Corrado, G.; Dean, J.\n          Efficient Estimation\nof Word Representations\nin Vector Space. 1st International Conference\non Learning Representations; ICLR 2013\ue5f8Workshop Track Proceedings, 2013."},{"key":"2026081804434806900_cit60","first-page":"10456","article-title":"Stable and\nExpressive Recurrent Vision Models","volume":"33","author":"Linsley","year":"2020","journal-title":"Adv. Neural\nInf. Process. Syst."},{"key":"2026081804434806900_cit61","doi-asserted-by":"publisher","first-page":"154","DOI":"10.1039\/c9sc04944d","article-title":"Datasets\nand Their Influence on the Development of Computer Assisted Synthesis\nPlanning Tools in the Pharmaceutical Domain","volume":"11","author":"Thakkar","year":"2020","journal-title":"Chem. Sci."},{"key":"2026081804434806900_cit62","doi-asserted-by":"crossref","unstructured":"Klein, G.; Kim, Y.; Deng, Y.; Senellart, J.; Rush, A.\n          OpenNMT: Open-Source Toolkit\nfor Neural Machine Translation. Proceedings\nof ACL 2017, System Demonstrations; Association\nfor Computational Linguistics: Stroudsburg,\nPA, USA, 2017; pp 67\u201372.","DOI":"10.18653\/v1\/P17-4012"},{"key":"2026081804434806900_cit63","unstructured":"Jin, W.; Coley, C.; Barzilay, R.; Jaakkola, T.\n          Predicting Organic Reaction\nOutcomes with Weisfeiler-Lehman Network. 31st\nConference on Neural Information Processing Systems (NIPS 2017), 2017."},{"key":"2026081804434806900_cit64","unstructured":"Bradshaw, J.; Kusner, M.; Paige, B.; Segler, M.; Herna\u0301ndez-Lobato, J.\n          Generative Model\nFor Electron Paths. ICLR: International Conference\non Learning Representations, 2019."},{"key":"2026081804434806900_cit65","doi-asserted-by":"publisher","first-page":"6091","DOI":"10.1039\/c8sc02339e","article-title":"\u201cFound\nin Translation\u201d: predicting outcomes of complex organic chemistry\nreactions using neural sequence-to-sequence models","volume":"9","author":"Schwaller","year":"2018","journal-title":"Chem. Sci."},{"key":"2026081804434806900_cit66","doi-asserted-by":"publisher","first-page":"1103","DOI":"10.1021\/acscentsci.7b00303","article-title":"Retrosynthetic Reaction\nPrediction Using Neural Sequence-to-Sequence\nModels","volume":"3","author":"Liu","year":"2017","journal-title":"ACS Cent. Sci."},{"key":"2026081804434806900_cit67","unstructured":"Tran, K.\n          \n          Optimization\nof\nMolecular Transformers: Influence of Tokenization Schemes. M.Sc. Thesis, Chalmers University of Technology, 2021."},{"key":"2026081804434806900_cit68","article-title":"Energy-Based View of Retrosynthesis","author":"Sun","year":"2020"},{"key":"2026081804434806900_cit69","doi-asserted-by":"publisher","first-page":"D1202","DOI":"10.1093\/nar\/gkv951","article-title":"PubChem Substance and Compound Databases","volume":"44","author":"Kim","year":"2016","journal-title":"Nucleic Acids Res."},{"key":"2026081804434806900_cit70","article-title":"ChemBERTa: Large-Scale\nSelf-Supervised Pretraining for Molecular Property Prediction","author":"Chithrananda","year":"2020"},{"key":"2026081804434806900_cit71","doi-asserted-by":"publisher","first-page":"513","DOI":"10.1039\/c7sc02664a","article-title":"MoleculeNet: A Benchmark\nfor Molecular Machine Learning","volume":"9","author":"Wu","year":"2018","journal-title":"Chem. Sci."},{"key":"2026081804434806900_cit72","doi-asserted-by":"publisher","first-page":"10793","DOI":"10.1021\/acs.jpclett.1c03058","article-title":"Extracting Predictive\nRepresentations from Hundreds of Millions of Molecules","volume":"12","author":"Chen","year":"2021","journal-title":"J. Phys. Chem. Lett."}],"container-title":["Journal of Chemical Information\nand Modeling"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/pubs.acs.org\/doi\/pdf\/10.1021\/acs.jcim.2c01407","content-type":"application\/pdf","content-version":"vor","intended-application":"unspecified"},{"URL":"https:\/\/pubs.acs.org\/jcisd8\/article-pdf\/63\/7\/1914\/8908919\/ci2c01407.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/pubs.acs.org\/jcisd8\/article-pdf\/63\/7\/1914\/8908919\/ci2c01407.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,18]],"date-time":"2026-08-18T09:49:03Z","timestamp":1787046543000},"score":1,"resource":{"primary":{"URL":"https:\/\/pubs.acs.org\/jcisd8\/article\/63\/7\/1914\/850309\/Transformer-Performance-for-Chemical-Reactions"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,3,23]]},"references-count":72,"journal-issue":{"issue":"7","published-online":{"date-parts":[[2023,3,23]]},"published-print":{"date-parts":[[2023,4,10]]}},"URL":"https:\/\/doi.org\/10.1021\/acs.jcim.2c01407","relation":{},"ISSN":["1549-9596","1549-960X"],"issn-type":[{"value":"1549-9596","type":"print"},{"value":"1549-960X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,3,23]]}}}