Mingliang Zhai
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Mingliang Zhai has a total of 12 co-authors. Here are the top co-authors by number of shared publications:
- Yunde Jia
School of Computer Science & Technology Beijing Key Laboratory of Intelligent Information Technology Beijing Institute of Technology China
3 shared publications - Yuwei Wu
Beijing Institute of Technology, Beijing, China and Shenzhen MSU-BIT University, Shenzhen, China
3 shared publications - Chenchen Jing
Zhejiang University of Technology
2 shared publications - Chen Yi
Department of Computer Vision Technology (VIS), Baidu Inc
1 shared publication - Chengquan Zhang
Baidu Inc
1 shared publication - Chuanhao Li
Shanghai Artificial Intelligence Laboratory
1 shared publication
Latest Publications
- 27 November 2024
- Published by Springer Nature
- p. 143-160
Compositional Substitutivity of Visual Reasoning for Visual Question Answering
- 27 November 2024
- Published by Springer Nature
- p. 143-160
Compositional Substitutivity of Visual Reasoning for Visual Question Answering
- 4 November 2024
- Published by Springer Nature
- p. 167-180
Visual-Guided Reasoning Path Generation for Visual Question Answering
- 4 November 2024
- Published by Springer Nature
- p. 167-180
Visual-Guided Reasoning Path Generation for Visual Question Answering
- Open Access
- 1 August 2023
- Published by International Joint Conferences on Artificial Intelligence
- Vol. 5, 5269-5277
Fast-StrucTexT: An Efficient Hourglass Transformer with Modality-guided Dynamic Token Merge for Document Understanding
Transformers achieve promising performance in document understanding because of their high effectiveness and still suffer from quadratic computational complexity dependency on the sequence length. General efficient transformers are challenging to be directly adapted to model document. They are unable to handle the layout representation in documents, e.g. word, line and paragraph, on different granularity levels and seem hard to achieve a good trade-off between efficiency and performance. To tackle the concerns, we propose Fast-StrucTexT, an efficient multi-modal framework based on the StrucTexT algorithm with an hourglass transformer architecture, for visual document understanding. Specifically, we design a modality-guided dynamic token merging block to make the model learn multi-granularity representation and prunes redundant tokens. Additionally, we present a multi-modal interaction module called Symmetry Cross-Attention (SCA) to consider multi-modal fusion and efficiently guide the token mergence. The SCA allows one modality input as query to calculate cross attention with another modality in a dual phase. Extensive experiments on FUNSD, SROIE, and CORD datasets demonstrate that our model achieves the state-of-the-art performance and almost 1.9x faster inference time than the state-of-the-art methods.
- 1 August 2023
- Published by International Joint Conferences on Artificial Intelligence
- Vol. 5, 5269-5277
Open AccessFast-StrucTexT: An Efficient Hourglass Transformer with Modality-guided Dynamic Token Merge for Document Understanding
Transformers achieve promising performance in document understanding because of their high effectiveness and still suffer from quadratic computational complexity dependency on the sequence length. General efficient transformers are challenging to be directly adapted to model document. They are unable to handle the layout representation in documents, e.g. word, line and paragraph, on different granularity levels and seem hard to achieve a good trade-off between efficiency and performance. To tackle the concerns, we propose Fast-StrucTexT, an efficient multi-modal framework based on the StrucTexT algorithm with an hourglass transformer architecture, for visual document understanding. Specifically, we design a modality-guided dynamic token merging block to make the model learn multi-granularity representation and prunes redundant tokens. Additionally, we present a multi-modal interaction module called Symmetry Cross-Attention (SCA) to consider multi-modal fusion and efficiently guide the token mergence. The SCA allows one modality input as query to calculate cross attention with another modality in a dual phase. Extensive experiments on FUNSD, SROIE, and CORD datasets demonstrate that our model achieves the state-of-the-art performance and almost 1.9x faster inference time than the state-of-the-art methods.