Saved in:
Bibliographic Details
Main Authors: Jia, Mengzhao, Yu, Wenhao, Ma, Kaixin, Fang, Tianqing, Zhang, Zhihan, Ouyang, Siru, Zhang, Hongming, Yu, Dong, Jiang, Meng
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2410.01744
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908395666145280
author Jia, Mengzhao
Yu, Wenhao
Ma, Kaixin
Fang, Tianqing
Zhang, Zhihan
Ouyang, Siru
Zhang, Hongming
Yu, Dong
Jiang, Meng
author_facet Jia, Mengzhao
Yu, Wenhao
Ma, Kaixin
Fang, Tianqing
Zhang, Zhihan
Ouyang, Siru
Zhang, Hongming
Yu, Dong
Jiang, Meng
contents Text-rich images, where text serves as the central visual element guiding the overall understanding, are prevalent in real-world applications, such as presentation slides, scanned documents, and webpage snapshots. Tasks involving multiple text-rich images are especially challenging, as they require not only understanding the content of individual images but reasoning about inter-relationships and logical flows across multiple visual inputs. Despite the importance of these scenarios, current multimodal large language models (MLLMs) struggle to handle such tasks due to two key challenges: (1) the scarcity of high-quality instruction tuning datasets for text-rich multi-image scenarios, and (2) the difficulty in balancing image resolution with visual feature sequence length. To address these challenges, we propose Leopard, an MLLM tailored for handling vision-language tasks involving multiple text-rich images. First, we curated about one million high-quality multimodal instruction-tuning data, tailored to text-rich, multi-image scenarios. Second, we proposed an adaptive high-resolution multi-image encoding module to dynamically optimize the allocation of visual sequence length based on the original aspect ratios and resolutions of images. Experiments on a diverse set of benchmarks reveal that our model consistently outperforms state-of-the-art systems, such as Llama-3.2 and Qwen2-VL, in challenging text-rich, multi-image evaluations. Remarkably, our approach achieves outstanding performance using only 1.2M training instances, all of which are fully open-sourced, demonstrating both high efficiency and effectiveness compared to models trained on large-scale in-house data. Our code and data are available at https://github.com/tencent-ailab/Leopard.
format Preprint
id arxiv_https___arxiv_org_abs_2410_01744
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Leopard: A Vision Language Model For Text-Rich Multi-Image Tasks
Jia, Mengzhao
Yu, Wenhao
Ma, Kaixin
Fang, Tianqing
Zhang, Zhihan
Ouyang, Siru
Zhang, Hongming
Yu, Dong
Jiang, Meng
Computer Vision and Pattern Recognition
Computation and Language
Text-rich images, where text serves as the central visual element guiding the overall understanding, are prevalent in real-world applications, such as presentation slides, scanned documents, and webpage snapshots. Tasks involving multiple text-rich images are especially challenging, as they require not only understanding the content of individual images but reasoning about inter-relationships and logical flows across multiple visual inputs. Despite the importance of these scenarios, current multimodal large language models (MLLMs) struggle to handle such tasks due to two key challenges: (1) the scarcity of high-quality instruction tuning datasets for text-rich multi-image scenarios, and (2) the difficulty in balancing image resolution with visual feature sequence length. To address these challenges, we propose Leopard, an MLLM tailored for handling vision-language tasks involving multiple text-rich images. First, we curated about one million high-quality multimodal instruction-tuning data, tailored to text-rich, multi-image scenarios. Second, we proposed an adaptive high-resolution multi-image encoding module to dynamically optimize the allocation of visual sequence length based on the original aspect ratios and resolutions of images. Experiments on a diverse set of benchmarks reveal that our model consistently outperforms state-of-the-art systems, such as Llama-3.2 and Qwen2-VL, in challenging text-rich, multi-image evaluations. Remarkably, our approach achieves outstanding performance using only 1.2M training instances, all of which are fully open-sourced, demonstrating both high efficiency and effectiveness compared to models trained on large-scale in-house data. Our code and data are available at https://github.com/tencent-ailab/Leopard.
title Leopard: A Vision Language Model For Text-Rich Multi-Image Tasks
topic Computer Vision and Pattern Recognition
Computation and Language
url https://arxiv.org/abs/2410.01744