Financial document understanding remains a critical challenge for Large Language Models, primarily due to the complex interplay between narrative text and structured numerical tables. Existing Retrieval-Augmented Generation (RAG) systems often treat...
Although multimodal Large Language Models (LLMs) perform well on document question answering tasks, Answer Correctness, i.e., whether a model’s answer is correct with respect to the question, does not fully describe model behavior because it do...
Question answering from visually rich documents (VRDs) is the task of retrieving the correct answer to a natural language question by considering the content of textual and visual elements in the document, as well as the pages’ layout. To answe...
The financial business process worldwide suffers from huge dependencies upon labor and written documents, thus making it tedious and time-consuming. In order to solve this problem, traditional robotic process automation (RPA) has recently been develo...
Technical documents differ from general text corpora in ways that complicate retrieval-augmented generation (RAG). Evidence for a single answer is often distributed across numbered clauses, tables, figures, captions, and ordered procedures rather tha...
This pilot study evaluates five LLM-based systems on newspaper front pages. Using ten front pages from Italian national daily newspapers, we assess 21 structured questions covering four capabilities: (i) bibliographic metadata extraction, (ii) layout...
Decades-old enterprise-resource-planning (ERP) systems lock operational data inside unstructured, human-readable reports, forcing slow, costly, error-prone manual re-keying. Because multimodal large language model (MLLM) capability is uneven, deployi...
Component obsolescence and supply-chain disruptions increasingly force engineers to spend significant time manually searching and comparing PDF datasheets to identify compatible replacement parts. We propose an AI-powered datasheet assistant based on...
Power system testing and inspection documents are multimodal and highly structured, making content-level audit labeling challenging due to scattered evidence and cross-component dependencies. We propose EviCal, an evidence-grounded consistency calibr...
Key information extraction (KIE) models are increasingly used in high-stakes document workflows, where each extracted value must be accompanied by evidence a reviewer can verify. Evaluating their explanations is difficult: human rationale annotation...
Spam detection has been a topic of extensive research; however, there has been limited focus on multimodal spam detection. In this study, we introduce a novel approach for multilingual multimodal spam detection, presenting the Multilingual and Multim...
This study develops a domain-adaptive multimodal RAG (Retrieval-Augmented Generation) system to improve the accuracy and efficiency of technical question answering based on large-scale structured manuals. Using Hyundai Staria maintenance documents as...
Cracks are a common defect in civil infrastructures, and their occurrence is often closely related to structural loading conditions, material properties, design and construction, and other factors. Therefore, detecting and analyzing cracks in civil i...
Multimodal Knowledge Graphs (MMKGs) offer a promising paradigm for integrating heterogeneous sources into a unified, queryable, semantically structured representation. However, existing MMKG construction pipelines remain predominantly text-centric, e...
A receipt information extraction task requires both textual and spatial analyses. Early receipt analysis systems primarily relied on template matching to extract data from spatially structured documents. However, these methods lack generalizability a...
A wide spectrum of digital data are becoming available to researchers and industries interested in the recording, documentation, recognition, and reproduction of human activities. In this work, we propose an approach for understanding and articulatin...
Recreating vanished architectural heritage while incorporating public accessibility remains a core challenge in cultural preservation. Here we present a unified multimodal framework that integrates evidence-based computational reconstruction with mul...
Visually rich document understanding (VrDU) models rely on tightly coupled textual, layout, and visual representations. In real-world deployments, these models must continuously adapt to new document domains over time. However, naïve sequential...
This study explores the evolving trends and research issues in the field of drone-assisted multimodal logistics over the past two decades. By employing various text-mining techniques on related research publications, we identify the most frequently i...
Paper-based logistics documents remain widely used in multi-enterprise supply chains, where heterogeneous layouts, noisy document images, non-standard material descriptions, and limited edge-computing resources make structured extraction and material...
Retrieval-Augmented Generation (RAG) has become a central approach to enhance the factual consistency and domain specificity of large language models (LLMs) by incorporating external context at inference time. However, most existing RAG systems rely...
Traditional ethnographic methods have long been employed to study craft practices, yet they often fall short of capturing the full depth of embodied knowledge, material interactions, and procedural workflows inherent in craftsmanship. This paper intr...
With the advent of big data science, the field of geoscience has undergone a paradigm shift toward data-driven scientific discovery. However, the abundance of geoscience data distributed across multiple sources poses significant challenges to researc...
Spontaneous duodenobiliary reflux is a rare condition in veterinary medicine, previously documented in only one canine case. We report a second instance in a 3-year-old neutered male Yorkshire Terrier, detected incidentally during a barium contrast s...
29 September 2026
Museums increasingly employ digital playful experiences to engage younger audiences, yet designing for both engagement and deeper historical understanding remains challenging. This paper presents Letters from the Past, a sequenced multimodal museum l...
Geotechnical hazards such as landslides, subsidence, slope instability, and infrastructure deformation threaten rapidly urbanising and environmentally stressed regions worldwide, intensifying the need for scalable and intelligent monitoring systems c...
Learning Design, as a field of research, provides practitioners with guidelines towards more effective teaching and learning. In parallel, observational methods (manual or automated) have been used in the classroom to reflect on and refine teaching a...
The manual creation of personalized travel itineraries remains a labor-intensive process that requires travel agents to consolidate heterogeneous information from multiple sources, including natural language interactions, booking confirmations, scree...
In the last two decades, forensic pathology and crime scene investigations have seen a rapid increase in examination tools due to the implementation of several imaging techniques, e.g., CT and MR scanning, surface scanning and photogrammetry. These t...
Objectives: To understand the clinical characteristics, risk factors, diagnosis, treatment, and outcomes of urinary tract obstruction caused by fungal balls. Materials and Methods: A comprehensive search was conducted across PubMed, Embase, and Medli...
With artificial intelligence (AI) rapidly increasing in popularity and presence in everyday life, new applications utilizing AI are being explored across virtually all domains, from banking and healthcare to cybersecurity to generative AI for images,...
Construction progress monitoring remains predominantly manual, labor-intensive, and reliant on subjective human interpretation. Human dependence often leads to redundant or unreliable information, resulting in scheduling delays and increased costs. A...
Background: Age-related decline in musculoskeletal function is a significant concern, particularly in Western countries facing demographic shifts and increased healthcare demands. This review examines the typical trajectories of musculoskeletal deter...
Background and Objectives: PAPSS2 loss-of-function variants cause autosomal recessive brachyolmia type 4 with skeletal dysplasia. Despite over 90 documented cases, molecular characterization remains incomplete, with many variants classified as uncert...
The deployment of large language models (LLMs) within the healthcare sector has sparked both enthusiasm and apprehension. These models exhibit the remarkable ability to provide proficient responses to free-text queries, demonstrating a nuanced unders...
The common methods used for grain-size analysis have their own deficiencies and limitations in terms of explaining the genesis of grain-size components. In this study, the end-member modeling analysis method is applied to multi-mode grain-size distri...
Breast cancer (BC) is the most common cancer type among women with a distinct clinical presentation, but the survival rate remains moderate despite advances in multimodal therapy. Consequently, a deeper understanding of the molecular etiology is requ...
Cetaceans are well known for their unique behavioral habits, such as calls and tactics. The possibility that these are acquired through social learning continues to be explored. This study investigates the ability of a young beluga whale to imitate n...
Arsenic, a naturally occurring metalloid, poses a significant global public health threat due to widespread environmental contamination. Despite its well-documented carcinogenicity, critical gaps remain in understanding the health impacts of chronic...
Background: Ultrasound-guided radiofrequency ablation (RFA) has become an established minimally invasive treatment for selected benign and autonomously functioning thyroid nodules. Although clinical outcomes are well documented, the biological mechan...
The transient receptor potential vanilloid type 1 (TRPV1) is a multimodal receptor which responds to various stimuli, including capsaicin, protons, and heat. Recent advances in cryo-electron microscopy have revealed the structures of TRPV1. However,...
Olive oil, a cornerstone of the Mediterranean diet, contains a saponifiable lipid fraction rich in oleic acid, and a non-saponifiable fraction composed of minor bioactive constituents such as squalene, vitamin E, oleuropein aglycone, hydroxytyrosol,...
Metadata extraction from natural history collection labels is a pivotal task for the online publication of digitized specimens. However, given the scale of these collections—which are estimated to host more than 2 billion specimens worldwide, i...
Pediatric extracorporeal membrane oxygenation (ECMO) is associated with complex hemostatic disturbances arising from interactions among blood–circuit interfaces, developmental hemostasis, endothelial dysfunction, coagulation, fibrinolysis, infl...
Facial Emotion Recognition (FER) has transformed human–computer interaction, yet its application to understanding relationships between physical environments and human emotional responses remains an emerging area within urban studies. This stud...
Sentiment analysis of online game reviews has increasingly become important in understanding player experiences and supporting data-driven game development. However, research in this domain has continuously faced two unresolved challenges: (1) the ex...
Background/Objectives: We present a case documenting the successful treatment for a patient with chronic low back pain (CLBP), chronic neck pain (CNP), and decreased quality of life improving after conservative therapy. CLBP has been the leading caus...
Background: Neurosarcoidosis is a highly variable condition with many clinical and radiological manifestations, that can lead to difficult identification of isolated central nervous system (CNS) forms, because it could mimic inflammatory, infective o...
Background: Acinic cell carcinoma (AciCC) comprises 6–7% of all salivary gland neoplasms and is the second most common salivary gland malignancy in children. Like many salivary gland carcinomas, it is considered low grade but occasionally it be...