{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,15]],"date-time":"2026-05-15T01:18:05Z","timestamp":1778807885895,"version":"3.51.4"},"reference-count":70,"publisher":"IEEE","license":[{"start":{"date-parts":[[2025,10,19]],"date-time":"2025-10-19T00:00:00Z","timestamp":1760832000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,10,19]],"date-time":"2025-10-19T00:00:00Z","timestamp":1760832000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,10,19]]},"DOI":"10.1109\/iccv51701.2025.02274","type":"proceedings-article","created":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T19:45:49Z","timestamp":1777491949000},"page":"24528-24538","source":"Crossref","is-referenced-by-count":1,"title":["MolParser: End-to-End Visual Recognition of Molecule Structures in the Wild"],"prefix":"10.1109","author":[{"given":"Xi","family":"Fang","sequence":"first","affiliation":[{"name":"DP Technology"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jiankun","family":"Wang","sequence":"additional","affiliation":[{"name":"DP Technology"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xiaochen","family":"Cai","sequence":"additional","affiliation":[{"name":"DP Technology"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Shangqian","family":"Che","sequence":"additional","affiliation":[{"name":"DP Technology"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Shuwen","family":"Yang","sequence":"additional","affiliation":[{"name":"DP Technology"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Haoyi","family":"Tao","sequence":"additional","affiliation":[{"name":"DP Technology"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Nan","family":"Wang","sequence":"additional","affiliation":[{"name":"DP Technology"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Lin","family":"Yao","sequence":"additional","affiliation":[{"name":"DP Technology"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Linfeng","family":"Zhang","sequence":"additional","affiliation":[{"name":"DP Technology"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Guolin","family":"Ke","sequence":"additional","affiliation":[{"name":"DP Technology"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1021\/acs.jcim.0c00042"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1145\/1553374.1553380"},{"key":"ref3","article-title":"Nougat: Neural optical understanding for academic documents","author":"Blecher","year":"2023","journal-title":"arXiv preprint"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/ICDAR.2019.00-41"},{"key":"ref5","article-title":"Img2smi: translating molecular structure images to simplified molecular-input line-entry system","author":"Campos","year":"2021","journal-title":"arXiv preprint"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1186\/s13321-024-00926-w"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1039\/D1SC01839F"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref9","volume-title":"Indigo library"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1038\/s42256-021-00438-4"},{"key":"ref11","article-title":"Unicorn: A unified contrastive learning approach for multi-view molecular representation learning","author":"Feng","year":"2024","journal-title":"arXiv preprint"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1021\/ci800067r"},{"key":"ref13","article-title":"Robust method of segmentation and recognition of chemical structure images in cheminfty","volume-title":"Pre-proceedings of the 9th IAPR international workshop on graphics recognition, GREC","author":"Fujiyoshi","year":"2011"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1007\/s44267-024-00067-6"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1093\/nar\/gkr777"},{"key":"ref16","first-page":"207","article-title":"Markush structure reconstruction","volume-title":"Gesellschaft f\u00fcr Informatik (GI) publishes this series in order to make available to a broad public recent findings in informatics (ie computer science and information systems), to document conferences that are organized in co-operation with GI and to publish the annual GI Award dissertation.","author":"Haupt","year":"2009"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1186\/1758-2946-5-7"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1145\/3534678.3539321"},{"key":"ref19","volume-title":"Bristol-myers squibb - molecular translation.","author":"Howard","year":"2021"},{"key":"ref20","article-title":"Strategies for pre-training graph neural networks","author":"Hu","year":"2019","journal-title":"arXiv preprint"},{"key":"ref21","article-title":"Gpt-4o system card","author":"Hurst","year":"2024","journal-title":"arXiv preprint"},{"key":"ref22","volume-title":"Ultralytics YOLO11","author":"Jocher","year":"2024"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1002\/cmtd.202100069"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1093\/nar\/gky1033"},{"key":"ref25","volume-title":"Rdkit: Open-source cheminformatics software.","author":"Landrum","year":"2023"},{"key":"ref26","first-page":"18893","article-title":"Pix2struct: Screenshot parsing as pretraining for visual language understanding","volume-title":"International Conference on Machine Learning","author":"Lee","year":"2023"},{"key":"ref27","article-title":"Bart: Denoising sequence-to-sequence pretraining for natural language generation, translation, and comprehension","author":"Lewis","year":"2019","journal-title":"arXiv preprint"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1021\/acs.jcim.3c02082"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v37i11.26538"},{"key":"ref30","first-page":"36","article-title":"Visual instruction tuning","author":"Liu","year":"2024","journal-title":"Advances in neural information processing systems"},{"key":"ref31","article-title":"Pre-training molecular graph representation with 3d geometry","author":"Liu","year":"2021","journal-title":"arXiv preprint"},{"key":"ref32","first-page":"21497","article-title":"A group symmetric stochastic differential equation model for molecule multi-modal pretraining","volume-title":"International Conference on Machine Learning","author":"Liu","year":"2023"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00986"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-main.966"},{"key":"ref35","first-page":"36","article-title":"Rethinking tokenizer and decoder in masked graph modeling for molecules","author":"Liu","year":"2024","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref36","article-title":"Omniparser for pure vision based gui agent","author":"Lu","year":"2024","journal-title":"arXiv preprint"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1021\/ci00008a018"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.01791"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.01352"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1021\/acs.jcim.0c00459"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1145\/1943403.1943444"},{"key":"ref42","volume-title":"Molvec v0.9.8","author":"Peryea","year":"2022"},{"key":"ref43","article-title":"Clef-ip 2010: Retrieval experiments in the intellectual property domain","volume-title":"CLEF (notebook papers\/labs\/workshops)","author":"Piroi","year":"2010"},{"key":"ref44","article-title":"Robust molecular image recognition: A graph generation approach","author":"Qian","year":"2022","journal-title":"Technical report, Technical Report"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1021\/acs.jcim.2c01480"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1186\/s13321-020-00469-w"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1186\/s13321-021-00538-8"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-023-40782-0"},{"key":"ref49","first-page":"12559","article-title":"Self-supervised graph transformer on large-scale molecular data","volume":"33","author":"Rong","year":"2020","journal-title":"Advances in neural information processing systems"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1117\/12.912185"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.6028\/nist.sp.500-296.chemical-gga"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1021\/acs.jcim.8b00669"},{"key":"ref53","first-page":"20479","article-title":"3d infomax improves gnns for molecular property prediction","volume-title":"International Conference on Machine Learning","author":"St\u00e4rk","year":"2022"},{"key":"ref54","article-title":"Large-scale representation learning on graphs via bootstrapping","author":"Thakoor","year":"2021","journal-title":"arXiv preprint"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1093\/bib\/bbac461"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1038\/s42256-022-00447-x"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1021\/ci00057a005"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1039\/D1SC02957F"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1039\/C7SC02664A"},{"key":"ref60","volume-title":"Molebert: Rethinking pre-training graph neural networks for molecules","author":"Xia","year":"2023"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.1007\/s11427-023-2388-x"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1021\/acs.jcim.2c00733"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.1186\/s13321-022-00624-5"},{"key":"ref64","first-page":"2024","article-title":"Mol-ae: Auto-encoder based molecular representation learning with 3d cloze test objective","author":"Yang","year":"2024","journal-title":"bioRxiv"},{"key":"ref65","doi-asserted-by":"publisher","DOI":"10.1093\/bioinformatics\/btac545"},{"key":"ref66","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9746088"},{"key":"ref67","article-title":"Multimodal molecular pre-training via modality blending","volume-title":"The Twelfth International Conference on Learning Representations","author":"Yu","year":"2024"},{"key":"ref68","doi-asserted-by":"publisher","DOI":"10.1039\/D4RA02442G"},{"key":"ref69","doi-asserted-by":"publisher","DOI":"10.1093\/bib\/bbac033"},{"key":"ref70","volume-title":"Uni-mol: A universal 3d molecular representation learning framework","author":"Zhou","year":"2023"}],"event":{"name":"2025 IEEE\/CVF International Conference on Computer Vision (ICCV)","location":"Honolulu, HI, USA","start":{"date-parts":[[2025,10,19]]},"end":{"date-parts":[[2025,10,25]]}},"container-title":["2025 IEEE\/CVF International Conference on Computer Vision (ICCV)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11443115\/11443287\/11445036.pdf?arnumber=11445036","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,30]],"date-time":"2026-04-30T06:19:56Z","timestamp":1777529996000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11445036\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,19]]},"references-count":70,"URL":"https:\/\/doi.org\/10.1109\/iccv51701.2025.02274","relation":{},"subject":[],"published":{"date-parts":[[2025,10,19]]}}}