GUIDE, a governed multi-agent framework built on a shared versioned rule store with schema-validated inter-agent contracts and end-to-end provenance tracking, is introduced, achieving 96% document success, and reduces turnaround to 40-125 minutes per document.
Abstract
Enterprise guideline documents are heterogeneous and multimodal, combining narrative text, complex tables, and embedded images. Existing LLM and VLM systems face hallucinated content, table structure degradation, and lack governed workflows extending beyond extraction to validation and artifact generation. This leaves enterprises to perform this manually, consuming 2-3 days per document. To address this, we introduce GUIDE, a governed multi-agent framework built on a shared versioned rule store with schema-validated inter-agent contracts and end-to-end provenance tracking. Six specialized agents handle parsing, VLM-driven extraction, consistency checking, evaluation, human-in-the-loop (HITL) escalation, and persona-tailored artifact synthesis. Evaluated on 120 real-world enterprise guideline documents, GUIDE achieves 96% document success, extracts 3,896 rules with 71.4% auto-approved, produces 812 deployment-ready artifacts, and reduces turnaround to 40-125 minutes per document.
LlamaExtract Agentic Plus ranks first on all three metrics, with accuracy comparable to coding agents at a fraction of the cost, and is the first to score value accuracy, record completeness at scale, grounding, and measured cost together.
Boyang Zhang, Adrian Lyjak, Elizabeth Stewart et al.· arXiv.org· 1 citation
Financial services enterprises process vast volumes of unstructured documents-loan agreements, compliance filings, audit reports, and customer correspondence-requiring synthesis, summarization, and regulatory cross-referencing under strict accuracy and traceability constraints. Conventional document automation relies on rule-based extraction or single-pass large language model (LLM) summarization, both of which struggle with provenance tracking, multi-document reasoning, and auditability demanded by financial regulators. This paper presents the Enterprise Generative Document Intelligence Platform (EGDIP), a layered system architecture integrating retrieval-augmented generation (RAG), microservices-based orchestration, and containerized deployment to deliver scalable, traceable document synthesis for regulated financial environments. EGDIP organizes processing into four cooperating layers: a data ingestion layer supporting heterogeneous document formats and streaming updates, a retrieval and indexing layer built on vector-based semantic search, an intelligence layer combining a fine-tuned LLM with deterministic compliance-rule validation, and a consumption layer exposing synthesized outputs through RESTful APIs and interactive dashboards. The system was deployed on a containerized Kubernetes environment and evaluated on a corpus of 180,000 financial documents from loan servicing and regulatory filing workflows. Results show a 58.3% reduction in document review time, a 39.7% improvement in cross-reference accuracy relative to keyword-based retrieval baselines, and sub-second retrieval latency at the evaluated scale. These findings demonstrate that combining retrieval-augmented generation with deterministic validation yields a practical, auditable path toward generative AI adoption in compliance-sensitive enterprise settings.
Arun Meesala· International Journal of Art...· 0 citations
The Assistant-Scribe-Knowledge Checker (ASK) framework for generating structured specification documents through guided interviews is evaluated in a consulting-firm setting where consultants are required to produce project “return-of-experience” documents to capture reusable knowledge.
Sylvain Roudiere, Bianca Lento· European Conference on Knowl...· 0 citations
The findings suggest that AI-based structured extraction may redefine how organisations formalise expertise, shifting from document-centric storage toward schema-driven knowledge architectures.
Dilyan Georgiev, E. Gourova· European Conference on Knowl...· 0 citations
Extracting reliable knowledge from unstructured materials literature remains a central bottleneck for data-driven and AI-enabled materials discovery. Large language models (LLMs) are reshaping this task by integrating multimodal document parsing, ontology-guided semantic grounding, structured extraction, and agentic verification into increasingly unified workflows. This review analyzes these developments through a Perception–Cognition–Action lens. At the perception layer, we examine how scientific document parsers, multimodal LLMs, table and chart readers, and optical chemical-structure-recognition systems convert visually rich papers into computable evidence. At the cognition layer, we discuss how ontologies and knowledge graphs constrain LLM outputs, support entity alignment, and reduce semantic ambiguity. At the action layer, we compare schema-based extraction, schema-free discovery, and agentic extraction as a control–coverage–autonomy spectrum rather than a simple succession of tools. We further argue that reliability is the decisive criterion for large-scale deployment, and synthesize failure modes, layered defenses, and evaluation protocols that connect source grounding, ontology constraints, physical verification, and human-in-the-loop review. By distinguishing demonstrated extraction capabilities from more speculative AI-scientist and self-driving-laboratory visions, this review provides a comparative and risk-aware account of how LLM-driven systems can produce evidence-linked, physically meaningful, and reusable materials knowledge.
Shuai Yang, Yi-Meng Wang, Qiong Tu et al.· Journal of Materials Informa...· 0 citations
