{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,6]],"date-time":"2026-02-06T06:56:53Z","timestamp":1770361013938,"version":"3.49.0"},"reference-count":44,"publisher":"MDPI AG","issue":"2","license":[{"start":{"date-parts":[[2026,2,5]],"date-time":"2026-02-05T00:00:00Z","timestamp":1770249600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["MAKE"],"abstract":"<jats:p>This study investigates how Vision Language Models (VLMs) can be used and methodically configured to extract Environmental, Social, and Governance (ESG) metrics from corporate sustainability reports, addressing the limitations of existing text-only and manual ESG data-extraction approaches. Using the Design Science Research Methodology, we developed an extraction artifact comprising a curated page-level dataset containing greenhouse gas (GHG) emission-reduction targets, an automated evaluation pipeline, model and text-preprocessing comparisons, and iterative prompt and few-shot refinement. Pages from oil and gas sustainability reports were processed directly by VLMs to preserve visual\u2013textual structure, enabling a controlled comparison of text, image, and combined input modalities, with extraction quality assessed at page and attribute level using F1-scores. Among tested models, Mistral Small 3.2 demonstrated the most stable performance and was used to evaluate image, text, and combined modalities. Combined text + image modality performed best (F1 = 0.82), particularly on complex page layouts. The findings demonstrate how to effectively integrate visual and textual cues for ESG metric extraction with VLMs, though challenges remain for visually dense layouts and avoiding inference-based hallucinations.<\/jats:p>","DOI":"10.3390\/make8020037","type":"journal-article","created":{"date-parts":[[2026,2,5]],"date-time":"2026-02-05T15:24:19Z","timestamp":1770305059000},"page":"37","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Enhancing the Extraction of GHG Emission-Reduction Targets from Sustainability Reports Using Vision Language Models"],"prefix":"10.3390","volume":"8","author":[{"ORCID":"https:\/\/orcid.org\/0009-0009-8500-2366","authenticated-orcid":false,"given":"Lars","family":"Wilhelmi","sequence":"first","affiliation":[{"name":"Faculty of Business and Economics, University of Goettingen, 37073 Goettingen, Germany"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-9570-1679","authenticated-orcid":false,"given":"Christian","family":"Bruns","sequence":"additional","affiliation":[{"name":"Faculty of Business and Economics, University of Goettingen, 37073 Goettingen, Germany"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2811-3096","authenticated-orcid":false,"given":"Matthias","family":"Schumann","sequence":"additional","affiliation":[{"name":"Faculty of Business and Economics, University of Goettingen, 37073 Goettingen, Germany"}]}],"member":"1968","published-online":{"date-parts":[[2026,2,5]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Cruz, C.A., and Matos, F. (2023). ESG Maturity: A Software Framework for the Challenges of ESG Data in Investment. Sustainability, 15.","DOI":"10.3390\/su15032610"},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Gutierrez-Bustamante, M., and Espinosa-Leal, L. (2022). Natural Language Processing Methods for Scoring Sustainability Reports\u2014A Study of Nordic Listed Companies. Sustainability, 14.","DOI":"10.20944\/preprints202207.0090.v1"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"1315","DOI":"10.1093\/rof\/rfac033","article-title":"Aggregate Confusion: The Divergence of ESG Ratings","volume":"26","author":"Berg","year":"2022","journal-title":"Rev. Financ."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"144572","DOI":"10.1016\/j.jclepro.2024.144572","article-title":"ESGReveal: An LLM-Based Approach for Extracting Structured Data from ESG Reports","volume":"489","author":"Zou","year":"2025","journal-title":"J. Clean. Prod."},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Escrig-Olmedo, E., Fern\u00e1ndez-Izquierdo, M., Ferrero-Ferrero, I., Rivera-Lirio, J., and Mu\u00f1oz-Torres, M. (2019). Rating the Raters: Evaluating How ESG Rating Agencies Integrate Sustainability Principles. Sustainability, 11.","DOI":"10.3390\/su11030915"},{"key":"ref_6","unstructured":"MSCI (2025). MSCI ESG Ratings, MSCI."},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Ni, J., Bingler, J., Colesanti Senni, C., Kraus, M., Gostlow, G., Schimanski, T., Stammbach, D., Vaghefi, S., Wang, Q., and Webersinke, N. (2023). chatReport: Democratizing Sustainability Disclosure Analysis through LLM-based Tools. arXiv.","DOI":"10.18653\/v1\/2023.emnlp-demo.3"},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Bedn\u00e1rov\u00e1, M., and Soratana, K. (2025). ESG Data and Metrics. Environmental, Social, and Governance (ESG) Investment and Reporting, Springer Nature Switzerland.","DOI":"10.1007\/978-3-031-84235-1"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"102720","DOI":"10.1016\/j.ribaf.2024.102720","article-title":"Greenwashing Prevention in Environmental, Social, and Governance (ESG) Disclosures: A Bibliometric Analysis","volume":"74","author":"Sneideriene","year":"2025","journal-title":"Res. Int. Bus. Financ."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1007\/s10551-022-05297-6","article-title":"Quantitative Research on Corporate Social Responsibility: A Quest for Relevance and Rigor in a Quickly Evolving, Turbulent World","volume":"187","author":"Du","year":"2022","journal-title":"J. Bus. Ethics"},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"1176","DOI":"10.1007\/s11142-021-09609-5","article-title":"Mandatory CSR and Sustainability Reporting: Economic Analysis and Literature Review","volume":"26","author":"Christensen","year":"2021","journal-title":"Rev. Acc. Stud."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"1497","DOI":"10.1038\/s41597-025-05664-8","article-title":"Addressing Data Gaps in Sustainability Reporting: A Benchmark Dataset for Greenhouse Gas Emission Extraction","volume":"12","author":"Beck","year":"2025","journal-title":"Sci. Data"},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Filipe, J., \u015amia\u0142ek, M., Brodsky, A., and Hammoudi, S. (2023). ESG Data Collection with Adaptive AI. Proceedings of the 25th International Conference on Enterprise Information Systems, SCITEPRESS\u2014Science and Technology Publications.","DOI":"10.1007\/978-3-031-64748-2"},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Hoffswell, J., and Liu, Z. (2019). Interactive Repair of Tables Extracted from PDF Documents on Mobile Devices. Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems, Association for Computing Machinery.","DOI":"10.1145\/3290605.3300523"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"41","DOI":"10.1140\/epjds\/s13688-024-00481-2","article-title":"Glitter or Gold? Deriving Structured Insights from Sustainability Reports via Large Language Models","volume":"13","author":"Bronzini","year":"2024","journal-title":"EPJ Data Sci."},{"key":"ref_16","unstructured":"Bordes, F., Pang, R.Y., Ajay, A., Li, A.C., Bardes, A., Petryk, S., Ma\u00f1as, O., Lin, Z., Mahmoud, A., and Jayaraman, B. (2024). An Introduction to Vision-Language Modeling. arXiv."},{"key":"ref_17","unstructured":"Peng, J., Gao, J., Tong, X., Guo, J., Yang, H., Qi, J., Li, R., Li, N., and Xu, M. (2024). Advanced Unstructured Data Processing for ESG Reports: A Methodology for Structured Transformation and Enhanced Analysis. arXiv."},{"key":"ref_18","unstructured":"Chen, C.-C., Huang, H.-H., Takamura, H., and Chen, H.-H. (2022). FinSim4-ESG Shared Task: Learning Semantic Similarities for the Financial Domain. Extended Edition to ESG Insights. Proceedings of the Fourth Workshop on Financial Technology and Natural Language Processing (FinNLP@IJCAI-ECAI 2022), Association for Computational Linguistics."},{"key":"ref_19","unstructured":"Wan, M., and Huang, C.-R. (2022). An NLP Approach for the Analysis of Global Reporting Initiative Indexes from Corporate Sustainability Reports. Proceedings of the LREC 2022 Workshop on The First Computing Social Responsibility Workshop: NLP Approaches to Corporate Social Responsibilities (CSR-NLP I 2022), European Language Resources Association (ELRA)."},{"key":"ref_20","unstructured":"Chen, C.-C., Liu, X., Hahn, U., Nourbakhsh, A., Ma, Z., Smiley, C., Hoste, V., Das, S.R., Li, M., and Ghassemi, M. (2024). NetZeroFacts: Two-Stage Emission Information Extraction from Company Reports. Proceedings of the Joint Workshop of the 7th Financial Technology and Natural Language Processing, the 5th Knowledge Discovery from Unstructured Data in Financial Services, and the 4th Workshop on Economics and Natural Language Processing, Association for Computational Linguistics."},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Schimanski, T., Bingler, J., Hyslop, C., Kraus, M., and Leippold, M. (2023). ClimateBERT-NetZero: Detecting and Assessing Net Zero and Reduction Targets. arXiv.","DOI":"10.2139\/ssrn.4599483"},{"key":"ref_22","unstructured":"Dave, A., Zhu, M., Hu, D., and Tiwari, S. (2024). Climate AI for Corporate Decarbonization Metrics Extraction. arXiv."},{"key":"ref_23","unstructured":"Adhikari, N.S., and Agarwal, S. (2024). A Comparative Study of PDF Parsing Tools Across Diverse Document Categories. arXiv."},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Huang, Y., Lv, T., Cui, L., Lu, Y., and Wei, F. (2022). LayoutLMv3: Pre-Training for Document AI with Unified Text and Image Masking. Proceedings of the 30th ACM International Conference on Multimedia, Association for Computing Machinery.","DOI":"10.1145\/3503161.3548112"},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Do, A.-D., and Do, T.-H. (2025). Adapting Vision-Language Models for Information Extraction from Bilingual Medical Invoices. Proceedings of the 2025 Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC), IEEE.","DOI":"10.1109\/APSIPAASC65261.2025.11249133"},{"key":"ref_26","unstructured":"Chiruzzo, L., Ritter, A., and Wang, L. (2025). MatViX: Multimodal Information Extraction from Visually Rich Articles. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers), Association for Computational Linguistics."},{"key":"ref_27","unstructured":"Drozd, A., Sedoc, J., Tafreshi, S., Akula, A., and Shu, R. (2025). Exploring Multimodal Language Models for Sustainability Disclosure Extraction: A Comparative Study. Proceedings of the Sixth Workshop on Insights from Negative Results in NLP, Association for Computational Linguistics."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3768156","article-title":"Large Language Models in Document Intelligence: A Comprehensive Survey, Recent Advances, Challenges, and Future Trends","volume":"44","author":"Ke","year":"2025","journal-title":"ACM Trans. Inf. Syst."},{"key":"ref_29","unstructured":"Yu, S., Tang, C., Xu, B., Cui, J., Ran, J., Yan, Y., Liu, Z., Wang, S., Han, X., and Liu, Z. (2025). VisRAG: Vision-Based Retrieval-Augmented Generation on Multi-Modality Documents. arXiv."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"45","DOI":"10.2753\/MIS0742-1222240302","article-title":"A Design Science Research Methodology for Information Systems Research","volume":"24","author":"Peffers","year":"2007","journal-title":"J. Manag. Inf. Syst."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"103502","DOI":"10.1016\/j.erss.2024.103502","article-title":"Greenwashing, Net-Zero, and the Oil Sands in Canada: The Case of Pathways Alliance","volume":"112","author":"Aronczyk","year":"2024","journal-title":"Energy Res. Soc. Sci."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3582688","article-title":"A Comprehensive Survey of Few-Shot Learning: Evolution, Applications, Challenges, and Opportunities","volume":"55","author":"Song","year":"2023","journal-title":"ACM Comput. Surv."},{"key":"ref_33","unstructured":"Kumar, A. (2023). Micro-Average, Macro-Average, Weighting: Precision, Recall, F1-Score. Anal. Yogi, Available online: https:\/\/vitalflux.com\/micro-average-macro-average-scoring-metrics-multi-class-classification-python\/."},{"key":"ref_34","unstructured":"Leung, K. (2022). Micro, Macro & Weighted Averages of F1 Score, Clearly Explained. Towards Data Sci., Available online: https:\/\/towardsdatascience.com\/micro-macro-weighted-averages-of-f1-score-clearly-explained-b603420b292f\/."},{"key":"ref_35","doi-asserted-by":"crossref","unstructured":"Foster, I., Ghani, R., Jarmin, R., Kreuter, F., and Lane, J. (2020). Big Data and Social Science, Chapman & Hall\/CRC. [2nd ed.].","DOI":"10.1201\/9780429324383"},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"J\u00fcnger, M., Liebling, T.M., Naddef, D., Nemhauser, G.L., Pulleyblank, W.R., Reinelt, G., Rinaldi, G., and Wolsey, L.A. (2010). The Hungarian Method for the Assignment Problem. 50 Years of Integer Programming 1958-2008: From the Early Years to the State-of-the-Art, Springer.","DOI":"10.1007\/978-3-540-68279-0"},{"key":"ref_37","unstructured":"Black, P.E. (2026, January 08). Greedy Algorithm. Dictionary of Algorithms and Data Structures 2005, Available online: https:\/\/xlinux.nist.gov\/dads\/HTML\/greedyalgo.html."},{"key":"ref_38","unstructured":"Poznanski, J., Rangapur, A., Borchardt, J., Dunkelberger, J., Huff, R., Lin, D., Rangapur, A., Wilhelm, C., Lo, K., and Soldaini, L. (2025). olmOCR: Unlocking Trillions of Tokens in PDFs with Vision Language Models. arXiv."},{"key":"ref_39","unstructured":"White, J., Fu, Q., Hays, S., Sandborn, M., Olea, C., Gilbert, H., Elnashar, A., Spencer-Smith, J., and Schmidt, D.C. (2023). A Prompt Pattern Catalog to Enhance Prompt Engineering with ChatGPT. Proceedings of the 30th Conference on Pattern Languages of Programs, The Hillside Group."},{"key":"ref_40","unstructured":"Yuan, J., Li, H., Ding, X., Xie, W., Li, Y.-J., Zhao, W., Wan, K., Shi, J., Hu, X., and Liu, Z. (2025, January 2\u20137). Understanding and Mitigating Numerical Sources of Nondeterminism in LLM Inference. Proceedings of the Thirty-ninth Annual Conference on Neural Information Processing Systems, San Diego, CA, USA."},{"key":"ref_41","doi-asserted-by":"crossref","unstructured":"Chen, L.-C., Weng, H.-T., Pardeshi, M.S., Chen, C.-M., Sheu, R.-K., and Pai, K.-C. (2025). Evaluation of Prompt Engineering on the Performance of a Large Language Model in Document Information Extraction. Electronics, 14.","DOI":"10.3390\/electronics14112145"},{"key":"ref_42","unstructured":"Che, W., Nabende, J., Shutova, E., and Pilehvar, M.T. (2025). Heuristic-Based Search Algorithm in Automatic Instruction-Focused Prompt Optimization: A Survey. Proceedings of the Findings of the Association for Computational Linguistics: ACL 2025, Association for Computational Linguistics."},{"key":"ref_43","doi-asserted-by":"crossref","unstructured":"Birti, M., Osborne, F., and Maurino, A. (2025, January 15\u201318). Optimizing Large Language Models for ESG Activity Detection in Financial Texts. Proceedings of the ICAIF \u201825: Proceedings of the 6th ACM International Conference on AI in Finance, Singapore.","DOI":"10.1145\/3768292.3770371"},{"key":"ref_44","doi-asserted-by":"crossref","unstructured":"Balke, W.-T., Golub, K., Manolopoulos, Y., Stefanidis, K., and Zhang, Z. (2026). ExtracTable: Human-in-the-Loop Transformation of Scientific Corpora into Structured Knowledge. Proceedings of the Linking Theory and Practice of Digital Libraries, Springer Nature Switzerland.","DOI":"10.1007\/978-3-032-05409-8"}],"container-title":["Machine Learning and Knowledge Extraction"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2504-4990\/8\/2\/37\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,2,5]],"date-time":"2026-02-05T15:53:49Z","timestamp":1770306829000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2504-4990\/8\/2\/37"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,2,5]]},"references-count":44,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2026,2]]}},"alternative-id":["make8020037"],"URL":"https:\/\/doi.org\/10.3390\/make8020037","relation":{},"ISSN":["2504-4990"],"issn-type":[{"value":"2504-4990","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,2,5]]}}}