TextHawk: Exploring Efficient Fine-Grained Perception of Multimodal Large Language Models

arXiv:2404.09204v1 Announce Type: new
Abstract: Multimodal Large Language Models (MLLMs) have shown impressive results on various multimodal tasks. However, most existing MLLMs are not well suited for document-oriented tasks, which require fine-grained image perception and information compression. In this paper, we present TextHawk, a MLLM that is specifically designed for document-oriented tasks, while preserving the general capabilities of MLLMs. TextHawk is aimed to explore efficient fine-grained perception by designing four dedicated components. Firstly, a ReSampling and ReArrangement (ReSA) module is …

abstract arxiv compression cs.ai cs.cv document fine-grained however image information language language models large language large language models mllm mllms multimodal paper perception results tasks type