https://ym-xu.github.io/living-survey/fields/document-understanding/#tag=kieKey information extraction from forms, invoices, receipts and records.Fri, 13 Nov 2026 14:11:24 +0000https://arxiv.org/abs/2610.26362https://arxiv.org/abs/2610.26362Mon, 19 Oct 2026 00:00:00 +0000Islam Mesabah, Ali Elhosary, Priyabanta Sandulu, Darko Obradovic, et al. Document annotation is a fundamental yet labor-intensive step in building document information extraction systems. While Optical Character Recognition (OCR) and Large Language Models (LLMs) can automate parts of this process, we argue that annotation systems should be judged not by how completely they remove human effo…https://arxiv.org/abs/2609.35796https://arxiv.org/abs/2609.35796Thu, 17 Sep 2026 00:00:00 +0000Xiem HoangVan, Phu TranQuang, Minh DinhBao, Tien VuHuu Invoices are commercial documents that contain various pieces of information, including the purchased items, time, and total money. Making the extraction of important information crucial. The stored information serves different purposes. Korean language is the native language of about 80 million people, playing an impo…https://arxiv.org/abs/2609.20110https://arxiv.org/abs/2609.20110Thu, 17 Sep 2026 00:00:00 +0000Yichao Jin, Yushuo Wang, Yuxuan Han, Kwan Ching Yee Sonia, et al. Straight-through processing (STP) on extracted key-value fields from financial documents without human review requires a calibrated probability together with a bounded guarantee on the residual error of the auto-approved tier. The emergence of modern Vision Language Models (VLMs) provides an out-of-the-box capability f…https://arxiv.org/abs/2609.15706https://arxiv.org/abs/2609.15706Mon, 14 Sep 2026 00:00:00 +0000Kushal Patel, Pushkal Shrivastava, Mackenzie Lees, Qirui Lu, et al. Vision-language models (VLMs) are increasingly used to extract structured fields from business documents, yet most evaluations report accuracy on clean benchmarks and offer little guidance to practitioners choosing an approach for a given task complexity. We address this gap with a measurement-grounded study and an ope…https://arxiv.org/abs/2608.12911https://arxiv.org/abs/2608.12911Thu, 13 Aug 2026 00:00:00 +0000Beining Xu, Hairui Wang, Jiaxin Wang, Changsheng Chen, et al. · ACM MM 2026 While the privacy risks of multimodal large language models (MLLMs) have drawn significant attention, the unique vulnerabilities of domain-specific MLLMs remain largely underexplored. Focusing on document understanding MLLMs for identity document processing, this paper investigates the privacy issues inherent in Key In…https://arxiv.org/abs/2608.28635https://arxiv.org/abs/2608.28635Sat, 08 Aug 2026 00:00:00 +0000Nimol Thuon, Panhapin Theang · ICPR 2026 Recent multimodal large language models (MLLMs) have advanced document understanding, visual question answering, and text extraction. However, their reliability in low-resource, non-Latin settings remains uncertain. Khmer form documents present particular challenges because they contain complex script forms, mixed Khme…https://arxiv.org/abs/2608.06607https://arxiv.org/abs/2608.06607Thu, 06 Aug 2026 00:00:00 +0000Sreerekha Rajendran Most document-extraction systems use a single model for all documents. This is simple but can be costly for easy cases and less effective for difficult ones. We examine whether we can predict a document's difficulty before extraction using inexpensive, document-based signals, and use this to choose between a cheaper an…https://arxiv.org/abs/2608.01792https://arxiv.org/abs/2608.01792Mon, 03 Aug 2026 00:00:00 +0000Priyashree Roy, Sujitha Martin, Mohammad Rostami, Spencer Romo, et al. Intelligent document processing (IDP) with vision-language models (VLMs) hinges on confidence scores trustworthy enough to route extractions between automation and human review. Existing document benchmarks are dominated by clean, high-quality samples, leaving low accuracy regions too sparse for calibration assessment.…https://arxiv.org/abs/2607.29677https://arxiv.org/abs/2607.29677Fri, 31 Jul 2026 00:00:00 +0000Boyang Zhang, Adrian Lyjak, Eli Stewart, Zhaoqi Li, et al. Enterprise workflows increasingly rely on agents for \emph{schema-guided extraction}: given a document and a user-defined schema, the agent faithfully follows the schema to produce the correct output with source evidence as grounding metadata. We present ExtractBench, a benchmark for schema-guided extraction and, to ou…https://arxiv.org/abs/2607.22723https://arxiv.org/abs/2607.22723Wed, 22 Jul 2026 00:00:00 +0000Huafu Li, Guo Chen, Jia Xia, Lei Wang, et al. Visual information extraction (VIE) from visually rich documents remains challenging due to high layout variability and real-world impairments. Existing methods typically rely on sequential OCR pipelines or end-to-end models requiring extensive labeled data and layout-specific training, limiting their scalability.We pr…https://arxiv.org/abs/2607.04636https://arxiv.org/abs/2607.04636Mon, 06 Jul 2026 00:00:00 +0000Zhipeng Xu, Zulong Chen, Qing Liu, Junhao Ji, et al. Key Information Extraction (KIE) converts visually rich documents into structured data, but practical deployment remains challenging: strong performance often relies on costly on-server Large Multimodal Models (LMMs), while compact locally deployable models lack sufficient KIE supervision. We present SAYRE, a scene-awa…https://arxiv.org/abs/2606.25343https://arxiv.org/abs/2606.25343Wed, 24 Jun 2026 00:00:00 +0000Heethanjan Kanagalingam, Thenukan Pathmanathan, Mokeeshan Vathanakumar, Basim Azam, et al. Vision Language Models have achieved near-human performance on single-document Visual Question Answering, yet their effectiveness degrades significantly when retrieving information from large collections of visually homogeneous documents. Existing multi-document benchmarks aggregate diverse document types, creating art…https://arxiv.org/abs/2606.24420https://arxiv.org/abs/2606.24420Tue, 23 Jun 2026 00:00:00 +0000Nitesh Kumar · IJCAI Workshop In high-stakes document processing pipelines, including financial reconciliation, compliance verification, and procurement automation, an LLM extraction that is silently wrong is more dangerous than one that is visibly absent. The central challenge is not extraction accuracy alone but reliable confidence estimation: kn…https://arxiv.org/abs/2605.22413https://arxiv.org/abs/2605.22413Thu, 21 May 2026 00:00:00 +0000Yandi Wang, Libin Zhan, Ziwei Huang, Tiancheng Luo, et al. · ACL 2026 Extracting structured information from visual documents (Visual Information Extraction, VIE) is a cornerstone of business automation. While recent Multimodal Large Language Models (MLLMs) have shown promising capabilities, existing benchmarks suffer from critical limitations in scale and realism, lack semantic granular…