Benchmarking Open-Source Layout Detection Models for Data Snapshot Extraction from Institutional Documents

Institutional documents contain substantial amounts of operational and analytical information embedded within figures and tables. Current approaches for extracting visual content from documents are largely built around generic document layout analysis, where figures and tables are treated as uniformly relevant document objects rather than semantically meaningful analytical artifacts. In this work, we introduce a benchmark dataset and evaluation framework for \textit{data snapshot extraction}, the task of identifying and localizing semantically meaningful visual artifacts within institutional documents. The benchmark spans humanitarian reports, World Bank policy research working papers, and project appraisal documents, and includes annotations for figures and tables that contain reusable analytical information. Using this dataset, we benchmarked multiple open-source layout detection models and evaluated both detection performance and spatial extraction quality. Our results show that current models struggle to generalize to operational institutional documents despite strong performance on conventional academic benchmarks. Common failure modes include confusion between analytical and non-analytical content, fragmentation of composite analytical artifacts, and incomplete extraction of contextual information required for interpretation. These findings highlight a persistent gap between generic document layout analysis and operationally useful data snapshot extraction. We release the source PDFs, annotation dataset, metadata, and source code to support future research in operational document intelligence. The dataset is available at https://huggingface.co/datasets/ai4data/data-snapshot and the source code is available at https://github.com/worldbank/ai4data/tree/main/experimental/data-snapshot.

End-to-End Text Line Detection and Ordering

Practical text-recognition pipelines for historical documents typically decompose layout analysis into line detection followed by a separate reading-order step, with the latter most often handled by a hand-coded geometric heuristic that struggles with marginalia, multiple columns, tables, and source-specific editorial conventions. This article introduces Orli (Ordered Regression of Lines), an end-to-end model that casts both sub-tasks as a single image-to-sequence problem: from a page image, Orli autoregressively generates text-line baselines directly in reading order. Baselines are represented in a chord-frame parameterization that anchors a line's position, orientation, and extent while encoding local geometry through perpendicular offsets; an iterative refinement head and a local visual refiner produce the final curve. Trained on a heterogeneous corpus of 196,691 pages spanning ten writing systems, Orli marginally exceeds the previously reported state of the art for cBAD line detection without dataset-specific training, reaches near perfect coverage and ordering on multiple reading-order benchmarks zero-shot, and adapts to more specialized out-of-domain layouts with limited fine-tuning. The method's source code and model weights are available under an open license at https://github.com/mittagessen/orli.

PereStruct: Multimodal Semantic Assembly for Robust Historical Document Parsing

Parsing historical documents with complex, non-standard layouts remains a fundamental bottleneck in large-scale archival digitization. Unlike modern typography, historical newspapers exhibit severe physical degradation and highly irregular page structures that confound even state-of-the-art vision-language models, presenting severe out-of-distribution challenges. We address this gap with an automated pipeline specifically designed for parsing historical newspapers, documents characterized by particularly intricate multi-column layouts. Our approach combines a fine-tuned YOLO architecture for layout analysis and block detection, trained on 1,426 fully human-annotated scanned pages, with a novel semantic assembly module that reconstructs articles by jointly modeling lexical-semantic similarity via TF-IDF, visual embeddings from our fine-tuned YOLO, and geometric layout constraints. This multi-modal integration yields state-of-the-art performance, achieving an F1 score of 0.904 on block-to-article mapping. Notably, end-to-end evaluation against vision-language models (Qwen3.6-35B-A3B and Qwen3.6-Plus) demonstrates that PereStruct achieves substantially higher fidelity (BLEU approximately 0.96 vs 0.34), validating that modular architectures excel where generic VLMs fail on complex historical layouts. To support reproducibility and advance research in this domain, we release both the training corpus of 599 annotated pages and a curated PereStruct benchmark of 93 pages with expert-verified ground-truth block-to-article mappings. This framework establishes a robust foundation for high-fidelity digitization and semantic reconstruction of complex archival materials.

Multimodal Approaches for Visually-Rich Document Type Classification: A Comparative Analysis

Document type classification in visually rich documents remains challenging, as relevant information is distributed across textual, visual, and layout modalities. To capture this complexity, current approaches rely on diverse multimodal modeling strategies, resulting in heterogeneous architectures that complicate systematic comparison. This variability is also reflected in existing comparative studies, which often rely on heterogeneous evaluation setups, further complicating systematic comparison and making it difficult to assess progress. To address these limitations, this work provides a structured analysis of multimodal design strategies across transformer- and LLM-based architectures, combined with a controlled empirical comparison within a unified experimental framework. Specifically, four representative models (LayoutLMv3, Donut, Qwen3-VL-32B-Instruct, and Qwen3-32B) are evaluated on the RVL-CDIP benchmark to systematically analyze the contributions of text, image, and layout information for document type classification, with a particular focus on contrasting OCR-dependent and OCR-free approaches. The results show that specialized multimodal Transformers outperform LLM-based approaches on visually rich and layout-intensive documents. Image information contributes most strongly to reliable classification, while OCR-derived text provides useful but secondary support. These findings highlight that multimodal processing remains essential for documents with pronounced layout structure. Overall, the study provides a systematic basis for comparing multimodal architectures and offers practical guidance for selecting effective feature combinations and model designs for document type classification.

Enginuity: A Dataset and Benchmark for Vision-Language Understanding of Engineering Diagrams

Engineering diagrams pose a distinct challenge for vision-language models: unlike natural images or general documents, they encode information through dense spatial layouts, domain-specific symbols, and cross-references between visual callouts and structured parts tables. Despite their centrality to service, repair, and design workflows, there is no public benchmark for measuring VLM capabilities in this domain; existing datasets primarily focus on flowcharts, scientific figures, or business documents. To address this gap, we introduce Enginuity, the first open dataset and benchmark for evaluating VLMs on complex engineering diagrams. We define two tasks over a corpus of U.S. military service and repair manuals: structured parts-table extraction (Task 1) and free-form visual diagram question answering (VQA)(Task 2) for benchmarking. We evaluate four frontier VLMs (GPT-5.2 Chat, Claude Opus 4.7, Gemma 4, Qwen3-VL-32B-Instruct) under zero-shot and chain-of-thought prompting. On Task 1, models reach Recall@all of 0.61-0.87 but Token F1pen of only 0.03-0.18, exposing a systematic gap between part identification and description fidelity. Task 2 reveals a consistent factual-reasoning gap across all models. A supporting analysis shows that token-overlap metrics under-report model capability on technical descriptions by 2-6x relative to semantic similarity, motivating LLM-as-judge calibration for domain-specific evaluation. We release the dataset, annotations, evaluation harness, and per-sample model outputs to support a reproducible study of VLM capability on engineering content.

Dr. DocBench: A Comprehensive Benchmark for Expert-Level and Difficult Document Parsing

Document parsing and recognition are fundamental capabilities for vision-language models (VLMs) and document processing systems. However, existing Optical Character Recognition (OCR) and document parsing benchmarks are increasingly limited in coverage and difficulty: many focus on common document genres or uniformly sampled pages where modern parsers already perform strongly, while offering limited annotation for expert-domain structures such as chemical formula, music notation, complex tables, and cross-page layouts. We introduce Dr. DocBench, a difficulty-aware benchmark for expert-level document parsing. Built from a large-scale multilingual book corpus, Dr. DocBench spans 52 BISAC subject domains and selects challenging documents through parser-failure-based sampling, targeting cases where multiple state-of-the-art systems struggle. It contains 4,514 annotated pages from long documents averaging around 100 pages, with 65k high-quality page- and block-level annotations for layout, reading order, hierarchical relations, and domain-specific visual contents. Evaluations of pipeline-based parsers and general-purpose VLMs show that strong performance on existing benchmarks does not transfer to our expert-level document parsing. Our analysis reveals substantial failures across subjects, content types, and structural attributes, highlighting Dr. DocBench as a comprehensive testbed for diagnosing and advancing document intelligence.

Can Retrieval Heads See Images? Multimodal Retrieval Heads in Long-Context Vision-Language Models

Large vision-language models increasingly rely on long-context modeling to reason over documents, hour-level videos, and long-horizon agent trajectories, requiring them to locate relevant evidence across interleaved text and images. Prior work has studied this behavior using retrieval heads in large language models, but its copy-based criterion does not directly apply when evidence appears in images. We introduce a multimodal retrieval head detection method that scores attention from question tokens to textual or visual evidence. With this method, we show that multimodal retrieval heads are sparse, intrinsic, and causally important: only 4.4-10.2% of attention heads account for 50% of the positive retrieval-score mass, and masking the top-5% selected heads drops MMLongBench-Doc from 48.2% to 5.7% and SlideVQA from 71.2% to 8.9%, while random-head masking is far less damaging. Further analysis shows that these heads are partly shared across modalities yet remain dynamic within each modality, with image retrieval heads changing more than text retrieval heads as context length and haystack modality change. Without further training, we find that these heads can also be used directly to rank visually rich documents: on MMDocIR, Qwen3-VL-8B selected-head scoring improves Recall@1 by 7.7/7.4 macro/micro points for page retrieval and 6.3/6.8 points for layout retrieval over the strongest reported baseline.

How Do Document Parsers Break? Auditing Structural Vulnerability in Document Intelligence

Document Layout Analysis (DLA) pipelines provide structured page representations for retrieval-augmented generation, long-document question answering, and other document intelligence systems, yet their robustness evaluation remains largely area-centric. We identify this Footprint Bias and propose a lightweight output-level auditing framework that decouples probe construction, policy-driven targeting, and structure-aware diagnosis. The framework combines Block-level Structural Loss Rate (B-SLR), granularity-aware exposure descriptors, and pathway attribution to analyze where perturbations interact with layout structure and how failures propagate. Across MinerU and PP-StructureV3 on 1,000 pages, affected area weakly tracks perturbation-induced OCR instability (R^2=0.384/0.110), whereas B-SLR aligns much more closely with it (R^2=0.727/0.916). Exposure descriptors further separate occlusion- and topology-dominant pathways, and small structurally targeted probes cause downstream QA/retrieval degradation comparable to larger-footprint perturbations. These results shift DLA robustness evaluation from footprint-based stress testing toward structure-aware vulnerability auditing.

Structured Layout Priors for Robust Out-of-Distribution Visual Document Understanding

Vision-Language Models (VLMs) parse documents end-to-end but frequently break down on layouts unlike those seen in training. We attribute this to a two-hop bottleneck: before the decoder can extract content (Hop 2), it must first classify and localize the enclosing layout entity (Hop 1), and when the first hop fails the second collapses into omissions, malformed structure, or autoregressive repetition. We pre-resolve Hop 1 outside the decoder by running a lightweight RT-DETR detector, serializing its outputs in the parser's native DocTags vocabulary, and injecting them into the prompt alongside the full page image. Unlike analyze-then-parse approaches that crop the page, or prior prompt-level priors written in plain text, our prior shares the decoder's generation space and leaves the global image in view as a fallback when detections are noisy. On a 10k-page structural out-of-distribution benchmark, markdown F1 rises from $0.37$ to $0.92$; on the Chinese subset of OmniDocBench, table TEDS rises from $0.01$ to $0.36$; and on the 26k-page ViDoRe V3 benchmark, infinite-loop decoding failures drop across every industrial domain tested. These gains cost $15\%$ wall-clock latency and a median of $74$ prompt tokens, with no architectural change to the base VLM. An attention-level analysis further reveals a bimodal phase shift in which the decoder attends to injected layout tokens when emitting structure and to image patches when emitting content, consistent with the two-hop bottleneck being alleviated. Model weights will be released to support reproducibility.

LLM-Augmented Semantic Steering of Text Embedding Projection Spaces

Low-dimensional projections of text embeddings support visual analysis of document collections, but their spatial organization may not reflect the relationships an analyst intends to examine. Existing semantic interaction approaches encode semantic intent indirectly through geometric constraints or model updates, limiting interpretability and flexibility. We introduce LLM-augmented semantic steering, which enables analysts to express semantic intent by grouping a small set of example documents within the projection. A large language model externalizes this intent as natural-language representations and selectively extends it to related documents; the resulting semantic information is then incorporated into document representations via text augmentation or embedding-level blending, without retraining the underlying models. A case study illustrates how the same corpus can be reorganized from different semantic perspectives, while simulation-based evaluation shows that semantic steering improves global and local alignment with target semantic structures using only minimal interaction. Embedding-level blending further enables continuous and controllable steering of projection layouts. These results position projection spaces as intent-dependent semantic workspaces that can be reshaped through explicit, interpretable, language-mediated interaction.