Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

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

About

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 proposed to reduce the redundancy in the document texts and lower the computational cost of the MLLM. We explore encoding the positions of each local feature by presenting Scalable Positional Embeddings (SPEs), which can preserve the scalability of various image sizes. A Query Proposal Network (QPN) is then adopted to initialize the queries dynamically among different sub-images. To further enhance the fine-grained visual perceptual ability of the MLLM, we design a Multi-Level Cross-Attention (MLCA) mechanism that captures the hierarchical structure and semantic relations of document images. Furthermore, we create a new instruction-tuning dataset for document-oriented tasks by enriching the multimodal document data with Gemini Pro. We conduct extensive experiments on both general and document-oriented MLLM benchmarks, and show that TextHawk outperforms the state-of-the-art methods, demonstrating its effectiveness and superiority in fine-grained document perception and general abilities.

Ya-Qi Yu, Minghui Liao, Jihao Wu, Yongxin Liao, Xiaoyu Zheng, Wei Zeng• 2024

Related benchmarks

TaskDatasetResultRank
Visual Question AnsweringChartQA--
519
OCR EvaluationOCRBench
Score76.4
350
Visual GroundingRefCOCO+ (val)
Accuracy86.2
253
Visual GroundingRefCOCO+ (testA)
Accuracy90
245
Visual GroundingRefCOCO+ (testB)
Accuracy80.4
219
Visual GroundingRefCOCO (testA)
Accuracy93
162
Visual GroundingRefCOCOg (val)
Accuracy88.2
158
Visual GroundingRefCOCOg (test)
Accuracy88.1
155
Infographic Question AnsweringInfoVQA
ANLS50.6
117
GroundingRefCOCO (val)
Accuracy91.9
23
Showing 10 of 17 rows

Other info

Follow for update