{"paper_id":"2307.11845","title":"Multimodal Document Analytics for Banking Process Automation","authors":["Christopher Gerling","Stefan Lessmann"],"paper_date":"2023-07-21","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2307.11845v2","page_url":"https://thequant.space/flowcharts/multimodal-document-analytics-for-banking-process-automation/","scores":{"empirical_rigor":8,"math_complexity":2.5,"quadrant":"Street Traders"},"proposes":{"status":"available","statement":"The paper proposes a robust multimodal document analysis method using LayoutXLM for banking document classification and demonstrates its effectiveness on German company register extracts.","quote":"we propose a robust multimodal document analysis method using LayoutXLM on German company register extracts.","summary":"Traditional banks face increasing competition from FinTechs in the rapidly evolving financial ecosystem. Raising operational efficiency is vital to address this challenge. Our study aims to improve the efficiency of document-intensive business processes in banking. To that end, we first review the…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Fine‑tuning is necessary because zero‑shot learning is not feasible for the token classification task.","type":"model","stated":"explicit","quote":"zero-shot learning is not feasible, and fine-tuning the model is necessary."},{"assumption":"Incorporating layout (visual) information substantially increases model performance.","type":"model","stated":"explicit","quote":"incorporating layout information in a model substantially increases its performance."}]},"datasets":{"status":"available","items":[{"name":"German company register extracts","provider":null,"kind":"text-news","access":"unknown","period":null,"universe":null,"quote":"German company register extracts"}],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"none"},"transaction_costs":{"status":"not_applicable","modeled":"n/a","headline_net_of_costs":null,"detail":null,"quote":null,"source":"front_matter","rule":"R3","candidates":{"front_matter":"n/a","evidence_card":"n/a","evidence_card_quoted":false},"meaning":"the paper reports no trading or execution result"},"results":{"status":"available","evaluation_type":"forecast-accuracy","findings":[{"finding":"After fine‑tuning for 10 epochs, LayoutXLM achieved an overall accuracy of 0.9758 on the validation dataset.","kind":"performance","quote":"This model achieved an overall accuracy of 0.9758 on the validation dataset."},{"finding":"LayoutXLM reaches macro‑average F1 scores of approximately 80% after convergence, outperforming text‑only variants.","kind":"performance","quote":"all three data types - text, layout, and image - shows the steepest learning curve and achieves the highest F1 scores of approx. 80% after convergence."},{"finding":"More than 75% of the best model performance (F1) can be achieved with as little as 30% of the training data.","kind":"performance","quote":"more than 75% of the best model performance (in terms of the F1 score) can be achieved with as little as 30% of the training data."},{"finding":"GPT‑3, used off‑the‑shelf, only attains an F1 score of 0.32, far below the fine‑tuned LayoutXLM model.","kind":"performance","quote":"it only achieves an F1 score of 0.32 (Figure 4)."},{"finding":"LayoutXLM can learn effectively from a minimal labeled set; 200‑300 observations (≈30% of data) raise F1 to about 70%, with a ceiling near 80% using the full set.","kind":"performance","quote":"... are sufficient for the model to effectively distinguish between the 10 token classes"}],"reported_performance":null,"out_of_sample":{"method":"holdout","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":38,"citations":{"cited_by_in_corpus":0,"by_independent_authors":0,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":0},"independent_papers":[],"scope":"Citations between papers in this corpus only; citations from outside it are not counted."}},"implementation":{"status":"available","public":false,"code_url":null,"note":"No repository presented by the authors as their own was found in the paper. Absence here is not proof that no code exists."},"replication":{"status":"none_found","card":null,"note":"No replication of this paper has been run by thequant.space yet."},"extraction":{"text_source":"fulltext","model":"openai/gpt-oss-120b","version":1,"extracted_at":"2026-11-14T18:24:18Z"},"provenance":{"claims":"Everything here is what the authors state in the paper. Nothing was re-run or reproduced by the platform.","quotes":"Assumptions, datasets, cost detail and findings were extracted by a language model. Every quote is verbatim paper text. Where an item has `support`, \"full\" means every figure in it appears in the quoted sentences and \"partial\" means a figure is stated elsewhere in the paper; items with a figure found nowhere in the paper were discarded.","datasets_known":"datasets.known lists well-known vendors and public datasets by a stable id, found by a dictionary over the full text where the sentence says the data is used; it is independent of the model and covers only names in the dictionary.","flags":"transaction_costs.modeled, results.evaluation_type and results.out_of_sample.method are platform classifications (structured-decision models); see https://thequant.space/score-guide/#penalty-flags","comparisons":"supports / contradicts / qualifies are a language model's reading of two papers' quoted findings, restricted to closely related papers that share no author. They are pointers to read, not verified replications, and coverage is partial.","implementation":"code_url is a repository the paper itself presents as its own (rule-scored, ambiguous cases judged by a model), checked for liveness when it was recorded."},"docs":"https://thequant.space/agents/"}