ThaiOCRBench: A Task-Diverse Benchmark for Vision-Language Understanding in Thai

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Nonesung, Surapon, Jaknamon, Teetouch, Chaiophat, Sirinya, Nitarach, Natapong, Wittayasakpan, Chanakan, Sirichotedumrong, Warit, Na-Thalang, Adisai, Pipatanakul, Kunat
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866912747777687552
author Nonesung, Surapon
Jaknamon, Teetouch
Chaiophat, Sirinya
Nitarach, Natapong
Wittayasakpan, Chanakan
Sirichotedumrong, Warit
Na-Thalang, Adisai
Pipatanakul, Kunat
author_facet Nonesung, Surapon
Jaknamon, Teetouch
Chaiophat, Sirinya
Nitarach, Natapong
Wittayasakpan, Chanakan
Sirichotedumrong, Warit
Na-Thalang, Adisai
Pipatanakul, Kunat
contents We present ThaiOCRBench, the first comprehensive benchmark for evaluating vision-language models (VLMs) on Thai text-rich visual understanding tasks. Despite recent progress in multimodal modeling, existing benchmarks predominantly focus on high-resource languages, leaving Thai underrepresented, especially in tasks requiring document structure understanding. ThaiOCRBench addresses this gap by offering a diverse, human-annotated dataset comprising 2,808 samples across 13 task categories. We evaluate a wide range of state-of-the-art VLMs in a zero-shot setting, spanning both proprietary and open-source systems. Results show a significant performance gap, with proprietary models (e.g., Gemini 2.5 Pro) outperforming open-source counterparts. Notably, fine-grained text recognition and handwritten content extraction exhibit the steepest performance drops among open-source models. Through detailed error analysis, we identify key challenges such as language bias, structural mismatch, and hallucinated content. ThaiOCRBench provides a standardized framework for assessing VLMs in low-resource, script-complex settings, and provides actionable insights for improving Thai-language document understanding.
format Preprint
id arxiv_https___arxiv_org_abs_2511_04479
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle ThaiOCRBench: A Task-Diverse Benchmark for Vision-Language Understanding in Thai
Nonesung, Surapon
Jaknamon, Teetouch
Chaiophat, Sirinya
Nitarach, Natapong
Wittayasakpan, Chanakan
Sirichotedumrong, Warit
Na-Thalang, Adisai
Pipatanakul, Kunat
Computation and Language
We present ThaiOCRBench, the first comprehensive benchmark for evaluating vision-language models (VLMs) on Thai text-rich visual understanding tasks. Despite recent progress in multimodal modeling, existing benchmarks predominantly focus on high-resource languages, leaving Thai underrepresented, especially in tasks requiring document structure understanding. ThaiOCRBench addresses this gap by offering a diverse, human-annotated dataset comprising 2,808 samples across 13 task categories. We evaluate a wide range of state-of-the-art VLMs in a zero-shot setting, spanning both proprietary and open-source systems. Results show a significant performance gap, with proprietary models (e.g., Gemini 2.5 Pro) outperforming open-source counterparts. Notably, fine-grained text recognition and handwritten content extraction exhibit the steepest performance drops among open-source models. Through detailed error analysis, we identify key challenges such as language bias, structural mismatch, and hallucinated content. ThaiOCRBench provides a standardized framework for assessing VLMs in low-resource, script-complex settings, and provides actionable insights for improving Thai-language document understanding.
title ThaiOCRBench: A Task-Diverse Benchmark for Vision-Language Understanding in Thai
topic Computation and Language
url https://arxiv.org/abs/2511.04479