A document is worth a structured record: Principled inductive bias design for document recognition

• URL: http://arxiv.org/abs/2507.08458v1
• Date: Fri, 11 Jul 2025 10:02:08 GMT
• Title: A document is worth a structured record: Principled inductive bias design for document recognition
• Authors: Benjamin Meyer, Lukas Tuggener, Sascha Hänzi, Daniel Schmid, Erdal Ayfer, Benjamin F. Grewe, Ahmed Abdulkadir, Thilo Stadelmann,
• Abstract summary: State-of-the-art approaches treat document recognition as a computer vision problem.<n>We suggest a novel perspective that frames document recognition as a transcription task from a document to a record.<n>This implies a natural grouping of documents based on the intrinsic structure inherent in their transcription.
• Score: 3.4332178437507936
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Many document types use intrinsic, convention-driven structures that serve to encode precise and structured information, such as the conventions governing engineering drawings. However, state-of-the-art approaches treat document recognition as a mere computer vision problem, neglecting these underlying document-type-specific structural properties, making them dependent on sub-optimal heuristic post-processing and rendering many less frequent or more complicated document types inaccessible to modern document recognition. We suggest a novel perspective that frames document recognition as a transcription task from a document to a record. This implies a natural grouping of documents based on the intrinsic structure inherent in their transcription, where related document types can be treated (and learned) similarly. We propose a method to design structure-specific inductive biases for the underlying machine-learned end-to-end document recognition systems, and a respective base transformer architecture that we successfully adapt to different structures. We demonstrate the effectiveness of the so-found inductive biases in extensive experiments with progressively complex record structures from monophonic sheet music, shape drawings, and simplified engineering drawings. By integrating an inductive bias for unrestricted graph structures, we train the first-ever successful end-to-end model to transcribe engineering drawings to their inherently interlinked information. Our approach is relevant to inform the design of document recognition systems for document types that are less well understood than standard OCR, OMR, etc., and serves as a guide to unify the design of future document foundation models.

Related papers

• DREAM: Document Reconstruction via End-to-end Autoregressive Model [53.51754520966657]
  We present an innovative autoregressive model specifically designed for document reconstruction, referred to as Document Reconstruction via End-to-end Autoregressive Model (DREAM)<n>We establish a standardized definition of the document reconstruction task, and introduce a novel Document Similarity Metric (DSM) and DocRec1K dataset for assessing the performance of the task.
  arXiv  Detail & Related papers  (2025-07-08T09:24:07Z)
• HDT: Hierarchical Document Transformer [70.2271469410557]
  HDT exploits document structure by introducing auxiliary anchor tokens and redesigning the attention mechanism into a sparse multi-level hierarchy. We develop a novel sparse attention kernel that considers the hierarchical structure of documents.
  arXiv  Detail & Related papers  (2024-07-11T09:28:04Z)
• Document Structure in Long Document Transformers [64.76981299465885]
  Long documents often exhibit structure with hierarchically organized elements of different functions, such as section headers and paragraphs. Despite the omnipresence of document structure, its role in natural language processing (NLP) remains opaque. Do long-document Transformer models acquire an internal representation of document structure during pre-training? How can structural information be communicated to a model after pre-training, and how does it influence downstream performance?
  arXiv  Detail & Related papers  (2024-01-31T08:28:06Z)
• Unified Pretraining Framework for Document Understanding [52.224359498792836]
  We present UDoc, a new unified pretraining framework for document understanding. UDoc is designed to support most document understanding tasks, extending the Transformer to take multimodal embeddings as input. An important feature of UDoc is that it learns a generic representation by making use of three self-supervised losses.
  arXiv  Detail & Related papers  (2022-04-22T21:47:04Z)
• Fourier Document Restoration for Robust Document Dewarping and Recognition [73.44057202891011]
  This paper presents FDRNet, a Fourier Document Restoration Network that can restore documents with different distortions. It dewarps documents by a flexible Thin-Plate Spline transformation which can handle various deformations effectively without requiring deformation annotations in training. It outperforms the state-of-the-art by large margins on both dewarping and text recognition tasks.
  arXiv  Detail & Related papers  (2022-03-18T12:39:31Z)
• Synthetic Document Generator for Annotation-free Layout Recognition [15.657295650492948]
  We describe a synthetic document generator that automatically produces realistic documents with labels for spatial positions, extents and categories of layout elements. We empirically illustrate that a deep layout detection model trained purely on the synthetic documents can match the performance of a model that uses real documents.
  arXiv  Detail & Related papers  (2021-11-11T01:58:44Z)
• Multilevel Text Alignment with Cross-Document Attention [59.76351805607481]
  Existing alignment methods operate at a single, predefined level. We propose a new learning approach that equips previously established hierarchical attention encoders for representing documents with a cross-document attention component.
  arXiv  Detail & Related papers  (2020-10-03T02:52:28Z)

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.