Aug 2026· International Journal of Creative and Open Research in Engineering and Management· Vol 02, pp. 1-9· 0 citations
TL;DR
DocuMind-AI is proposed, an end-to-end autonomous agentic architecture for document processing and intelligent question answering orchestrated via LangGraph that introduces a dynamic routing agent that assesses character density metrics to intelligently dispatch inputs between fast native text extractors and multimodal vision large language models.
Abstract
Document digitization and knowledge extraction remain challenging when dealing with heterogeneous PDF repositories comprising both machine-readable text and degraded, scanned visual artifacts. Traditional Optical Character Recognition (OCR) systems enforce rigid linear pipelines, while conventional Retrieval-Augmented Generation (RAG) models suffer from hallucination when context is sparse or noisy. In this paper, we propose DocuMind-AI, an end-to-end autonomous agentic architecture for document processing and intelligent question answering orchestrated via LangGraph. The framework introduces a dynamic routing agent that assesses character density metrics to intelligently dispatch inputs between fast native text extractors and multimodal vision large language models. Extracted text is normalized into structured Markdown through an automated cleansing agent and ingested into a dual-engine hybrid retrieval index combining BM25 lexical search with dense vector embeddings via Reciprocal Rank Fusion (RRF). Furthermore, a Corrective RAG (CRAG) self-reflection loop audits retrieved document chunks for semantic relevance, triggering automated query reformulation when retrieval confidence is low, and performs secondary hallucination auditing on synthesized answers. Empirical benchmarks demonstrate that our adaptive routing reduces multimodal API overhead by 68.4% on mixed corpora while achieving a 94.2% answer grounding accuracy, outperforming conventional single-engine RAG pipelines in both precision and computational efficiency.
Keywords— Agentic AI; Optical Character Recognition; Corrective RAG; LangGraph; Multimodal LLM; Hybrid Search.
Retrieval-Augmented Generation (RAG) has established itself as a compelling strategy for grounding large language model outputs in documentary evidence. However, production deployments continue to rely almost exclusively on homogeneous text corpora, even as enterprise repositories grow increasingly heterogeneous blending technical schematics, radiological images, annotated diagrams, and unstructured prose within the same archival system. This mismatch between system design and data reality motivates the present work. We propose a Multimodal RAG framework that unifies text and image retrieval through four tightly coupled components: a dual-stream embedding engine, a learned four-class AI query router, an adaptive confidence threshold, and a session-aware context store. Text is encoded with the allmpnet-base-v2 Sentence Transformer; images are embedded in the same 768-dimensional space via a domain-adapted CLIP ViT-L/14 model fine-tuned on approximately 120,000 technical and clinical text-image pairs. Concatenating 768-d vectors yields a 1,536-d composite query that drives a single approximate nearest-neighbor (ANN) search simultaneously across both modalities. A fine-tuned DistilBERT router assigns each query to one of four retrieval pathways text-only, image-only, hybrid, or conversational at 91.3% accuracy. Experiments were run on three enterprise corpora (engineering manuals, clinical case summaries, and legal paperwork) with the results being a mean F1@5 of 0.90, an increase of 22% over a dense text-only baseline, and a reduction of 49% in the number of hallucinations. The median first-token latency is 1.34s, which meets the interactive-use target deployment-contexts. These results show that it is possible to implement modality-aware retrieval in a simple and realistic setting with real-world organizational constraints and that the results are reproducible.
E.Vijayakumar, Ganesh A· 2026 4th International Confe...· 0 citations
Results demonstrate a 157% improvement in retrieval quality over a Naive-RAG baseline, with only 50 ms additional la tency, while Qwen2-VL-2B-Instruct achieved results comparable to cloud-based models in BERTScore, validate that open-source optimized SLMs, paired with advanced retrieval strategies, can provide competitive performance for document understanding without relying on cloud-based models.
HIEVI-RAG is introduced, a hierarchical, evidence-driven multimodal RAG framework for closed-domain document understanding that significantly outperforms existing open-source baselines and exceeds the strongest reported baseline by an average of 8.05% in accuracy.
Junyu Xiong, Yonghui Wang, Rongjian Gu et al.· 1 citation
Experimental results show that SAC-RAG reduces token consumption by 38%–58% at the cost of only a 1–2 percentage point EM drop, with EM actually improving after compression for reasoning-type questions, achieving the optimal quality–efficiency trade-off in terms of token consumption.
Deyu Zhang, Hongqiang Yu, Jinze Huo et al.· IEEE Access· 0 citations
Due to the surge in digital document creation, intelligent systems that can effectively retrieve accurate information from unstructured PDF content have become a necessity. Current keyword-based search techniques lack semantic meaning, while large language models (LLMs), when utilized individually, produce hallucinated results when they are not contextually aware of document content. In this paper, we introduce a novel real-time question answering system for PDF content using a lightweight approach to Retrieval-Augmented Generation (RAG). Our system combines a HuggingFace sentence transformer model for semantic embeddings with a FAISS vector similarity search for efficient retrieval, leveraging a Groq-hosted LLaMA 3.3 70B model for context-grounded answer generation. The system is built using LangChain for orchestration and Streamlit for web-based deployment, and requires no GPU, no fine-tuning, and no pre-indexed corpus. Evaluation across five complementary metrics—Exact Match, Token-level F1, BLEU, ROUGE, and semantic similarity—demonstrates strong alignment between generated and reference answers, with a peak semantic similarity of 0.8709 and a mean Token F1 of 0.4701, confirming the viability of this approach for real-world document intelligence.
Mukesh Lakshmi Sai Medikonda, Kalva Vishnu Teja, K. Greeshma et al.· International Conference Com...· 0 citations
While Multimodal Large Language Models (MLLMs) exhibit strong capabilities in document retrieval, their zero-shot performance is frequently bottlenecked by a static inference paradigm. Once deployed, these frozen models struggle to dynamically adjust their representation spaces to accommodate highly diverse query semantics and intricate visual layouts. To overcome this inherent rigidity, we introduce Dynamic Adapter Refinement at Test-time (DART), a novel inference-stage adaptation strategy. Rather than fine-tuning the massive backbone or applying destructive global feature projections, DART injects a lightweight, residual Adapter network to re-rank the initial top-K candidates. Our method extracts soft pseudo-labels directly from the model's initial retrieval confidence to drive a self-supervised contrastive learning objective. Crucially, to prevent catastrophic representation drift and preserve the rich vision-language alignment acquired during pre-training, DART constrains the Adapter's parameter updates using a targeted 𝐿𝐿2regularization penalty. Empirical evaluations across a diverse suite of multimodal document retrieval benchmarks reveal that DART achieves consistent and significant gains in ranking precision. Furthermore, this dynamic refinement process introduces minimal computational latency, offering a highly efficient, plug-and-play solution for adaptive document retrieval.
Jing Zhang, Yaowei Wang, Chongyu Wang et al.· International Conference on...· 0 citations