Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Zhou, Yinan, Chen, Yuxin, Lin, Haokun, Wu, Yichen, Yang, Shuyu, Qi, Zhongang, Ma, Chen, Zhu, Li, Shan, Ying
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2411.17125
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866911094068477952
author Zhou, Yinan
Chen, Yuxin
Lin, Haokun
Wu, Yichen
Yang, Shuyu
Qi, Zhongang
Ma, Chen
Zhu, Li
Shan, Ying
author_facet Zhou, Yinan
Chen, Yuxin
Lin, Haokun
Wu, Yichen
Yang, Shuyu
Qi, Zhongang
Ma, Chen
Zhu, Li
Shan, Ying
contents With recent advances in Multimodal Large Language Models (MLLMs), grounding and referring capabilities have gained increasing attention for achieving detailed understanding and flexible user interaction. However, these capabilities still remain underdeveloped in visual document understanding due to the scarcity of fine-grained datasets and comprehensive benchmarks. To fill this gap, we propose the DOcument Grounding and Referring data engine (DOGR-Engine), which generates two types of high-quality fine-grained document data: (1) multi-granular parsing data to improve text localization and recognition, and (2) instruction-tuning data to activate MLLMs' grounding and referring capabilities in dialogue and reasoning. Using the DOGR-Engine, we construct DOGR-Bench, a benchmark covering seven grounding and referring tasks across three document types (chart, poster, and PDF document), offering a comprehensive evaluation of fine-grained document understanding. Leveraging the generated data, we further develop DOGR, a strong baseline model that excels in text localization and recognition, while precisely grounds and refers to key textual information during conversation and reasoning, thereby advancing document understanding to a finer granularity and enable flexible interaction paradigms.
format Preprint
id arxiv_https___arxiv_org_abs_2411_17125
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle DOGR: Towards Versatile Visual Document Grounding and Referring
Zhou, Yinan
Chen, Yuxin
Lin, Haokun
Wu, Yichen
Yang, Shuyu
Qi, Zhongang
Ma, Chen
Zhu, Li
Shan, Ying
Computer Vision and Pattern Recognition
Artificial Intelligence
With recent advances in Multimodal Large Language Models (MLLMs), grounding and referring capabilities have gained increasing attention for achieving detailed understanding and flexible user interaction. However, these capabilities still remain underdeveloped in visual document understanding due to the scarcity of fine-grained datasets and comprehensive benchmarks. To fill this gap, we propose the DOcument Grounding and Referring data engine (DOGR-Engine), which generates two types of high-quality fine-grained document data: (1) multi-granular parsing data to improve text localization and recognition, and (2) instruction-tuning data to activate MLLMs' grounding and referring capabilities in dialogue and reasoning. Using the DOGR-Engine, we construct DOGR-Bench, a benchmark covering seven grounding and referring tasks across three document types (chart, poster, and PDF document), offering a comprehensive evaluation of fine-grained document understanding. Leveraging the generated data, we further develop DOGR, a strong baseline model that excels in text localization and recognition, while precisely grounds and refers to key textual information during conversation and reasoning, thereby advancing document understanding to a finer granularity and enable flexible interaction paradigms.
title DOGR: Towards Versatile Visual Document Grounding and Referring
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2411.17125