Reading Order Inference for Complex Document Layouts

• URL: http://arxiv.org/abs/2607.01018v1
• Date: Wed, 01 Jul 2026 14:52:00 GMT
• Title: Reading Order Inference for Complex Document Layouts
• Abstract summary: We present a method to evaluate the accuracy of non-modelled text-as-a-report.<n>We compare our method to the canonical Glossainaria layout in non-rectangular, non-modelled text.
• Score: 0.784573461502689
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Reading order inference remains a critical bottleneck in the digitization of complex historical manuscripts, where pages contain multiple spatially interleaved reading streams, the canonical example being the Glossa Ordinaria layout, in which a central text is surrounded by commentaries that wrap around it in non-rectangular, non-convex regions. We present a training-free, graph-based framework: each OCR text line becomes a node in a directed candidate-transition graph, edges are scored by a weighted additive ensemble of two lightweight language-model signals (causal language model conditional likelihood and BERT next-sentence prediction, NSP; a third sentence-embedding signal was evaluated but did not improve reading order), and the global reading order is recovered as a degree-constrained directed path cover. To avoid the cascading "edge-theft" failures of greedy edge selection, we propose a max-regret inference rule that prioritizes commitments with high opportunity cost. We evaluate on synthetic Glossa Ordinaria grid layouts, on 23 ALTO page geometries (10 historical source pages plus mirrored and flipped variants), and on a 140-page multi-column English subset of OmniDocBench, comparing our method against the canonical recursive XY-cut (PaddleOCR PP-StructureV3) and two LayoutReader variants (layout-only and text+layout) on identical inputs. On wrap-around Glossa layouts our method recovers 95% of ground-truth successor edges on average vs. XY-cut's 50%; on the OmniDocBench multi-column subset it reaches 88% macro edge accuracy versus XY-cut's 75% and LayoutReader's 25%. The LayoutReader baselines transfer poorly due to a word-level vs. line-level granularity mismatch. We additionally verify mirror-invariance under horizontal and vertical page reflections: Our method changes by less than 1 percentage point, classical XY-cut by 2 points, and LayoutReader-T by up to 8 points.

Related papers

• Rethinking Handwritten Character Recognition [0.0]
  GraphemeNet is a unified multi-script architecture governed by two binary axes.<n>A script-specific asymmetric convolution injects a stroke scaffold as a weighted residual at every stage.<n>A Stroke Topology Module (STM) selects between global average pooling with gated fusion and cross-scale attention.<n>A Linear Capsule Routing (LCR) with $O(n)$ routing is shared universally.
  arXiv  Detail & Related papers  (2026-09-02T16:19:14Z)
• Codebook Agent: Amortized Topology Design for LLM Multi-Agent Systems [67.04448659688579]
  We develop a query-independent 16-entry codebook that ranks the top decoded candidates in a single batched forward pass.<n>With no iterative search and no message passing at test time, Codebook Agent is the most accurate method on all six benchmarks.
  arXiv  Detail & Related papers  (2026-09-02T08:10:22Z)
• A Multi-Framework Comparison of Outline Stages in Long-Form Generation with LLMs [9.826721433645526]
  Long-form generation exposes fundamental limitations of large language models.<n>SuperWriter ranks first in the length-constrained single-chapter mode.<n> follow-up experiments will expand the sample size and add cross-model evaluators.
  arXiv  Detail & Related papers  (2026-08-13T15:44:03Z)
• PaDoc: Layout-Grounded Parallel Decoding for Document Parsing [45.6402658645855]
  Document serializes page layouts and regional contents into one autoregressive sequence.<n>We propose PaDoc, a layout-grounded that treats the predicted layout as a branching structure over a shared page representation.<n>On a 384-page subset and one A800, PaDoc is the fastest end-to-end at five levels, improving valid-page throughput by 67.4-118%.
  arXiv  Detail & Related papers  (2026-08-06T15:16:18Z)
• One Patch Is Enough: Reinforcement-Optimized Visual Token Grounding for MLLM-Based Scene Text Spotting [51.09550363815034]
  Scene text spotting requires high-precision alignment between textual recognition and spatial localization.<n>We propose Single-Patch Text Spotting (SPaTS), a vision-centric framework that routes each text instance through a single anchor visual token.<n>We show that SPaTS consistently and significantly outperforms both frontier closed-source MLLMs and OCR MLLMs.
  arXiv  Detail & Related papers  (2026-07-30T09:17:48Z)
• End-to-End Text Line Detection and Ordering [0.0]
  Orli is an end-to-end model that casts both sub-tasks as a single image-to-sequence problem.<n>From a page image, Orli autoregressively generates text-line baselines directly in reading order.<n>Trained on a heterogeneous corpus of 196,691 pages spanning ten writing systems.
  arXiv  Detail & Related papers  (2026-06-02T19:29:32Z)
• Scaling Multi-Hop Training Data via Graph-Constrained Path Selection [40.133031562641044]
  We decouple the two operations: reasoning paths are enumerated offline over a graph of contextual keyword centroids, and the teacher is invoked only to pre-validated paths.<n>At equal training scale, constrained and unconstrained chains yield indistinguishable downstream performance, and the gain at full scale comes from a 4.4$times$ expansion of the usable corpus.<n>Fine-tuning Qwen3-32B on 80K examples constructed from the CUAD legal contract corpus improves closed-book Token F1 from 21.66% to 38.58%.
  arXiv  Detail & Related papers  (2026-05-29T12:39:03Z)
• $β$-CLIP: Text-Conditioned Contrastive Learning for Multi-Granular Vision-Language Alignment [53.42377319350806]
  $$-CLIP is a multi-granular text-conditioned contrastive learning framework.<n>$$-CAL addresses the semantic overlap inherent in this hierarchy.<n>$$-CLIP establishes a robust, adaptive baseline for dense vision-language correspondence.
  arXiv  Detail & Related papers  (2025-12-14T13:03:20Z)
• Focus Anywhere for Fine-grained Multi-page Document Understanding [24.76897786595502]
  This paper proposes Fox, an effective pipeline, hybrid data, and tuning strategy, that catalyzes LVLMs to focus anywhere on single/multi-page documents. We employ multiple vision vocabularies to extract visual hybrid knowledge for interleaved document pages. We render cross-vocabulary vision data as the foreground to achieve a full reaction of multiple visual vocabularies and in-document figure understanding.
  arXiv  Detail & Related papers  (2024-05-23T08:15:49Z)
• Text Reading Order in Uncontrolled Conditions by Sparse Graph Segmentation [71.40119152422295]
  We propose a lightweight, scalable and generalizable approach to identify text reading order. The model is language-agnostic and runs effectively across multi-language datasets. It is small enough to be deployed on virtually any platform including mobile devices.
  arXiv  Detail & Related papers  (2023-05-04T06:21:00Z)
• End-to-End Page-Level Assessment of Handwritten Text Recognition [69.55992406968495]
  HTR systems increasingly face the end-to-end page-level transcription of a document. Standard metrics do not take into account the inconsistencies that might appear. We propose a two-fold evaluation, where the transcription accuracy and the RO goodness are considered separately.
  arXiv  Detail & Related papers  (2023-01-14T15:43:07Z)
• Span Pointer Networks for Non-Autoregressive Task-Oriented Semantic Parsing [55.97957664897004]
  An effective recipe for building seq2seq, non-autoregressive, task-orienteds to map utterances to semantic frames proceeds in three steps. These models are typically bottlenecked by length prediction. In our work, we propose non-autoregressives which shift the decoding task from text generation to span prediction.
  arXiv  Detail & Related papers  (2021-04-15T07:02:35Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.