{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,23]],"date-time":"2026-06-23T13:50:10Z","timestamp":1782222610049,"version":"3.54.5"},"reference-count":48,"publisher":"Wiley","issue":"2","license":[{"start":{"date-parts":[[2026,3,8]],"date-time":"2026-03-08T00:00:00Z","timestamp":1772928000000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"},{"start":{"date-parts":[[2026,3,8]],"date-time":"2026-03-08T00:00:00Z","timestamp":1772928000000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Transactions in GIS"],"published-print":{"date-parts":[[2026,4]]},"abstract":"<jats:title>ABSTRACT<\/jats:title>\n                  <jats:p>The geological reports and maps accumulated during geological surveying and mapping harbor rich expert knowledge and metallogenic clues. However, efficiently integrating and mining structured knowledge from complex multimodal data of polymetallic deposits remains a critical bottleneck in intelligent mineral prediction. To address this, we propose a knowledge graph (KG)\u2010enhanced multimodal retrieval\u2010augmented generation (RAG) framework, Geo\u2010MAG, for geological map understanding. Specifically, the framework first processes textual geological reports and constructs a structured KG. Concurrently, a vision large model parses geological maps to extract metadata, including legends, geological structures, strata, and lithologies. Leveraging this metadata, relevant subgraphs are retrieved from the KG to facilitate text\u2013map semantic alignment and enhance background geological knowledge. Finally, the integrated map information and structured subgraphs of KG are fed into the GPT\u20104o to enable deep semantic interpretation. Experimental results demonstrate that integrating the knowledge graph significantly boosts the GPT\u20104o's reasoning capability and interpretability in geological map understanding. The model achieves 77.2% accuracy in geological reasoning tasks, outperforming the direct end\u2010to\u2010end GPT\u20104o interpretation by 53.7% and lightweight schemes on the basis of basic metadata by 37.4%. This work represents a pioneering application of KG and RAG in geological map understanding, highlighting the synergistic advantages of integrating text and maps, and offering a novel perspective on multimodal integration within the geoscience domain.<\/jats:p>","DOI":"10.1111\/tgis.70226","type":"journal-article","created":{"date-parts":[[2026,3,9]],"date-time":"2026-03-09T06:05:32Z","timestamp":1773036332000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Geo\u2010\n                    <scp>MAG<\/scp>\n                    : A Knowledge Graph (\n                    <scp>KG<\/scp>\n                    )\u2010enhanced Multimodal Retrieval\u2010Augmented Generation (\n                    <scp>RAG<\/scp>\n                    ) Framework for Geological Map Understanding"],"prefix":"10.1111","volume":"30","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-5432-1166","authenticated-orcid":false,"given":"Kai","family":"Ma","sequence":"first","affiliation":[{"name":"Hubei Key Laboratory of Intelligent Vision Based Monitoring for Hydroelectric Engineering China Three Gorges University  Yichang China"},{"name":"College of Computer and Information Technology China Three Gorges University  Yichang China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Qingzhong","family":"Zhan","sequence":"additional","affiliation":[{"name":"Hubei Key Laboratory of Intelligent Vision Based Monitoring for Hydroelectric Engineering China Three Gorges University  Yichang China"},{"name":"College of Computer and Information Technology China Three Gorges University  Yichang China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tongyun","family":"Zhang","sequence":"additional","affiliation":[{"name":"Geological and Geographic Information Institute of Hunan Province\/Geological Big Data Center of Hunan Province  Changsha China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yuzheng","family":"Shi","sequence":"additional","affiliation":[{"name":"Geological and Geographic Information Institute of Hunan Province\/Geological Big Data Center of Hunan Province  Changsha China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhenxi","family":"Fang","sequence":"additional","affiliation":[{"name":"Key Laboratory of Metallogenic Prediction of Nonferrous Metals and Geological Environment Monitoring (Ministry of Education) \/ School of Geosciences and Info\u2010Physics Central South University  Changsha China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Syed Yasir Ali","family":"Shah","sequence":"additional","affiliation":[{"name":"Key Laboratory of Metallogenic Prediction of Nonferrous Metals and Geological Environment Monitoring (Ministry of Education) \/ School of Geosciences and Info\u2010Physics Central South University  Changsha China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Or Aimon Brou Koffi","family":"Kablan","sequence":"additional","affiliation":[{"name":"Key Laboratory of Metallogenic Prediction of Nonferrous Metals and Geological Environment Monitoring (Ministry of Education) \/ School of Geosciences and Info\u2010Physics Central South University  Changsha China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6075-9359","authenticated-orcid":false,"given":"Baoyi","family":"Zhang","sequence":"additional","affiliation":[{"name":"Key Laboratory of Metallogenic Prediction of Nonferrous Metals and Geological Environment Monitoring (Ministry of Education) \/ School of Geosciences and Info\u2010Physics Central South University  Changsha China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2026,3,8]]},"reference":[{"key":"e_1_2_11_2_1","doi-asserted-by":"publisher","DOI":"10.52202\/068431-1723"},{"key":"e_1_2_11_3_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.procs.2024.09.178"},{"key":"e_1_2_11_4_1","article-title":"Qwen\u2010Vl: A Versatile Vision\u2010Language Model for Understanding, Localization","volume":"2","author":"Bai J.","year":"2023","journal-title":"Text Reading, and Beyond"},{"key":"e_1_2_11_5_1","doi-asserted-by":"publisher","DOI":"10.3301\/IJG.2024.01"},{"key":"e_1_2_11_6_1","doi-asserted-by":"publisher","DOI":"10.1016\/S0169-7552(98)00110-X"},{"key":"e_1_2_11_7_1","article-title":"Bge m3\u2010Embedding: Multi\u2010Lingual, Multi\u2010Functionality, Multi\u2010Granularity Text Embeddings Through Self\u2010Knowledge Distillation","author":"Chen J.","year":"2024","journal-title":"arXiv Preprint arXiv:2402.03216"},{"key":"e_1_2_11_8_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.gsf.2022.101412"},{"key":"e_1_2_11_9_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.isprsjprs.2025.04.015"},{"key":"e_1_2_11_10_1","doi-asserted-by":"publisher","DOI":"10.1017\/S1351324924000044"},{"key":"e_1_2_11_11_1","doi-asserted-by":"publisher","DOI":"10.1007\/s42461-024-01114-y"},{"key":"e_1_2_11_12_1","doi-asserted-by":"publisher","DOI":"10.1145\/1327452.1327492"},{"key":"e_1_2_11_13_1","unstructured":"Edge D. H.Trinh N.Cheng et\u00a0al.2024.\u201cFrom Local to Global: A Graph Rag Approach to Query\u2010Focused Summarization. arXiv Preprint arXiv:2404.16130.\u201d"},{"key":"e_1_2_11_14_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.oregeorev.2025.106638"},{"key":"e_1_2_11_15_1","unstructured":"Guu K. K.Lee Z.Tung et\u00a0al.2020.\u201cRetrieval Augmented Language Model Pre\u2010training. International Conference on Machine Learning. PMLR 2020: 3929\u20133938.\u201d"},{"key":"e_1_2_11_16_1","doi-asserted-by":"crossref","unstructured":"He H. andJ. D.Choi.2021.\u201cThe Stem Cell Hypothesis: Dilemma Behind Multi\u2010Task Learning With Transformer Encoders. arXiv Preprint arXiv:2109.06939.\u201d","DOI":"10.18653\/v1\/2021.emnlp-main.451"},{"key":"e_1_2_11_17_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.array.2025.100399"},{"key":"e_1_2_11_18_1","doi-asserted-by":"publisher","DOI":"10.1002\/gdj3.245"},{"key":"e_1_2_11_19_1","doi-asserted-by":"crossref","unstructured":"Huang Y. T.Gao H.Xu et\u00a0al.2025.\u201cPeace: Empowering Geologic Map Holistic Understanding with mllms. Proceedings of the Computer Vision and Pattern Recognition Conference. 2025: 3899\u20133908.\u201d","DOI":"10.1109\/CVPR52734.2025.00369"},{"key":"e_1_2_11_20_1","doi-asserted-by":"publisher","DOI":"10.1109\/TBDATA.2019.2921572"},{"key":"e_1_2_11_21_1","first-page":"9459","article-title":"Retrieval\u2010Augmented Generation for Knowledge\u2010Intensive Nlp Tasks","volume":"33","author":"Lewis P.","year":"2020","journal-title":"Advances in Neural Information Processing Systems"},{"key":"e_1_2_11_22_1","doi-asserted-by":"publisher","DOI":"10.1007\/s11430-023-1444-9"},{"key":"e_1_2_11_23_1","doi-asserted-by":"publisher","DOI":"10.1080\/17538947.2025.2510566"},{"key":"e_1_2_11_24_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.acags.2025.100233"},{"key":"e_1_2_11_25_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jag.2025.104712"},{"key":"e_1_2_11_26_1","unstructured":"Lin C.\u2010Y.2004.\u201cRouge: A Package for Automatic Evaluation of Summaries. Text Summarization Branches Out. 2004: 74\u201381.\u201d"},{"key":"e_1_2_11_27_1","unstructured":"Ling C. X.Zhao J.Lu et\u00a0al.2023.\u201cDomain Specialization as the Key to Make Large Language Models Disruptive: A Comprehensive Survey. arXiv 2305.18703.\u201d"},{"key":"e_1_2_11_28_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.cageo.2022.105082"},{"key":"e_1_2_11_29_1","unstructured":"Manvi R. S.Khanna G.Mai et\u00a0al.2024.\u201cGEOLLM: EXTRACTING GEOSPATIAL KNOWLEDGE FROM LARGE LANGUAGE MODELS[C]\/\/12th International Conference on Learning Representations ICLR 2024 May 7 2024 \u2013 May 11 2024. Hybrid Vienna Austria:International Conference on Learning Representations ICLR 2024: Google Deepmind; Google Research; Meta; Microsoft.\u201d"},{"key":"e_1_2_11_30_1","doi-asserted-by":"publisher","DOI":"10.1007\/s12145-024-01463-8"},{"key":"e_1_2_11_31_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.asr.2024.09.013"},{"key":"e_1_2_11_32_1","doi-asserted-by":"crossref","unstructured":"Peng W. G.Li Y.Jiang et\u00a0al.2024.\u201cLarge Language Model Based Long\u2010Tail Query Rewriting in Taobao Search. 33rd Companion of the ACM World Wide Web Conference WWW 2023 May 13 2024 \u2013 May 17 2024. Singapore Singapore: Association for Computing Machinery Inc 2024: 20\u201328.\u201d","DOI":"10.1145\/3589335.3648298"},{"key":"e_1_2_11_33_1","doi-asserted-by":"publisher","DOI":"10.1080\/20964471.2025.2483541"},{"key":"e_1_2_11_34_1","unstructured":"Radford A. J. W.Kim C.Hallacy et\u00a0al.2021.\u201cLearning Transferable Visual Models From Natural Language Supervision. International Conference on Machine Learning. PmLR 2021: 8748\u20138763.\u201d"},{"key":"e_1_2_11_35_1","doi-asserted-by":"crossref","unstructured":"Singhal R. P.Patwa P.Patwa et\u00a0al.2024.\u201cEvidence\u2010Backed Fact Checking Using RAG and Few\u2010Shot in\u2010Context Learning With LLMs. 7th Fact Extraction and VERification Workshop FEVER 2024 Miami FL United States:Association for Computational Linguistics (ACL) 2024: 91\u201398.\u201d","DOI":"10.18653\/v1\/2024.fever-1.10"},{"key":"e_1_2_11_36_1","unstructured":"TEAM GEMINI R.Anil S.Borgeaud et\u00a0al.2023.\u201cGemini: A Family of Highly Capable Multimodal Models. arXiv Preprint arXiv:2312.11805.\u201d"},{"key":"e_1_2_11_37_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-019-41695-z"},{"key":"e_1_2_11_38_1","doi-asserted-by":"publisher","DOI":"10.32604\/cmc.2024.054360"},{"key":"e_1_2_11_39_1","doi-asserted-by":"crossref","first-page":"107984","DOI":"10.52202\/079017-3429","article-title":"Yolov10: Real\u2010Time End\u2010To\u2010End Object Detection","volume":"37","author":"Wang A.","year":"2024","journal-title":"Advances in Neural Information Processing Systems"},{"key":"e_1_2_11_40_1","unstructured":"Wang B. C.Xu X.Zhao et\u00a0al.2024.\u201cMineru: An Open\u2010Source Solution for Precise Document Content Extraction. arXiv Preprint arXiv:2409.18839.\u201d"},{"key":"e_1_2_11_41_1","doi-asserted-by":"publisher","DOI":"10.1093\/nsr\/nwab027"},{"key":"e_1_2_11_42_1","doi-asserted-by":"publisher","DOI":"10.1080\/17538947.2025.2521786"},{"key":"e_1_2_11_43_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.oregeorev.2023.105651"},{"key":"e_1_2_11_44_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.oregeorev.2025.106514"},{"key":"e_1_2_11_45_1","doi-asserted-by":"publisher","DOI":"10.1080\/10095020.2025.2514177"},{"key":"e_1_2_11_46_1","doi-asserted-by":"publisher","DOI":"10.1109\/TGRS.2024.3493455"},{"key":"e_1_2_11_47_1","doi-asserted-by":"publisher","DOI":"10.1109\/MGRS.2024.3496478"},{"key":"e_1_2_11_48_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jag.2024.103976"},{"key":"e_1_2_11_49_1","doi-asserted-by":"publisher","DOI":"10.1007\/s11430-022-1169-9"}],"container-title":["Transactions in GIS"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1111\/tgis.70226","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1111\/tgis.70226","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1111\/tgis.70226","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,23]],"date-time":"2026-06-23T13:05:58Z","timestamp":1782219958000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1111\/tgis.70226"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,8]]},"references-count":48,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2026,4]]}},"alternative-id":["10.1111\/tgis.70226"],"URL":"https:\/\/doi.org\/10.1111\/tgis.70226","archive":["Portico"],"relation":{},"ISSN":["1361-1682","1467-9671"],"issn-type":[{"value":"1361-1682","type":"print"},{"value":"1467-9671","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3,8]]},"assertion":[{"value":"2025-12-25","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-02-20","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-03-08","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e70226"}}