ExStrucTiny: A Benchmark for Schema-Variable Structured Information Extraction from Document Images
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912900918018048 |
|---|---|
| author | Sibue, Mathieu Garza, Andres Muñoz Mensah, Samuel Shetty, Pranav Ma, Zhiqiang Liu, Xiaomo Veloso, Manuela |
| author_facet | Sibue, Mathieu Garza, Andres Muñoz Mensah, Samuel Shetty, Pranav Ma, Zhiqiang Liu, Xiaomo Veloso, Manuela |
| contents | Enterprise documents, such as forms and reports, embed critical information for downstream applications like data archiving, automated workflows, and analytics. Although generalist Vision Language Models (VLMs) perform well on established document understanding benchmarks, their ability to conduct holistic, fine-grained structured extraction across diverse document types and flexible schemas is not well studied. Existing Key Entity Extraction (KEE), Relation Extraction (RE), and Visual Question Answering (VQA) datasets are limited by narrow entity ontologies, simple queries, or homogeneous document types, often overlooking the need for adaptable and structured extraction. To address these gaps, we introduce ExStrucTiny, a new benchmark dataset for structured Information Extraction (IE) from document images, unifying aspects of KEE, RE, and VQA. Built through a novel pipeline combining manual and synthetic human-validated samples, ExStrucTiny covers more varied document types and extraction scenarios. We analyze open and closed VLMs on this benchmark, highlighting challenges such as schema adaptation, query under-specification, and answer localization. We hope our work provides a bedrock for improving generalist models for structured IE in documents. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_12203 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | ExStrucTiny: A Benchmark for Schema-Variable Structured Information Extraction from Document Images Sibue, Mathieu Garza, Andres Muñoz Mensah, Samuel Shetty, Pranav Ma, Zhiqiang Liu, Xiaomo Veloso, Manuela Computation and Language Enterprise documents, such as forms and reports, embed critical information for downstream applications like data archiving, automated workflows, and analytics. Although generalist Vision Language Models (VLMs) perform well on established document understanding benchmarks, their ability to conduct holistic, fine-grained structured extraction across diverse document types and flexible schemas is not well studied. Existing Key Entity Extraction (KEE), Relation Extraction (RE), and Visual Question Answering (VQA) datasets are limited by narrow entity ontologies, simple queries, or homogeneous document types, often overlooking the need for adaptable and structured extraction. To address these gaps, we introduce ExStrucTiny, a new benchmark dataset for structured Information Extraction (IE) from document images, unifying aspects of KEE, RE, and VQA. Built through a novel pipeline combining manual and synthetic human-validated samples, ExStrucTiny covers more varied document types and extraction scenarios. We analyze open and closed VLMs on this benchmark, highlighting challenges such as schema adaptation, query under-specification, and answer localization. We hope our work provides a bedrock for improving generalist models for structured IE in documents. |
| title | ExStrucTiny: A Benchmark for Schema-Variable Structured Information Extraction from Document Images |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2602.12203 |