77 citations · 205 across the 9 of their papers we have counts for
10 papers
Diffusion-based Document Layout Generation
Liu He, Yijuan Lu, John Corring +2
We develop a diffusion-based approach for various document layout sequence generation. Layout sequences specify the contents of a document design in an explicit format. Our novel d…
Unifying Vision, Text, and Layout for Universal Document Processing
Zineng Tang, Ziyi Yang, Guoxin Wang +6
We propose Universal Document Processing (UDOP), a foundation Document AI model which unifies text, image, and layout modalities together with varied task formats, including docume…
Understanding Long Documents with Different Position-Aware Attentions
Hai Pham, Guoxin Wang, Yijuan Lu +2
Despite several successes in document understanding, the practical task for long document understanding is largely under-explored due to several challenges in computation and how t…
Improving Structured Text Recognition with Regular Expression Biasing
Baoguang Shi, Wenfeng Cheng, Yijuan Lu +2
We study the problem of recognizing structured text, i.e. text that follows certain formats, and propose to improve the recognition accuracy of structured text by specifying regula…
TrOCR: Transformer-based Optical Character Recognition with Pre-trained Models
Minghao Li, Tengchao Lv, Jingye Chen +6
Text recognition is a long-standing research problem for document digitalization. Existing approaches are usually built based on CNN for image understanding and RNN for char-level…
LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding
Yang Xu, Yiheng Xu, Tengchao Lv +9
Pre-training of text and layout has proved effective in a variety of visually-rich document understanding tasks due to its effective model architecture and the advantage of large-s…
Renofeation: A Simple Transfer Learning Method for Improved Adversarial Robustness
Ting-Wu Chin, Cha Zhang, Diana Marculescu
Fine-tuning through knowledge transfer from a pre-trained model on a large-scale dataset is a widely spread approach to effectively build models on small-scale datasets. In this wo…
Towards Efficient Model Compression via Learned Global Ranking
Ting-Wu Chin, Ruizhou Ding, Cha Zhang +1
Pruning convolutional filters has demonstrated its effectiveness in compressing ConvNets. Prior art in filter pruning requires users to specify a target model complexity (e.g., mod…
Layer-compensated Pruning for Resource-constrained Convolutional Neural Networks
Ting-Wu Chin, Cha Zhang, Diana Marculescu
Resource-efficient convolution neural networks enable not only the intelligence on edge devices but also opportunities in system-level optimization such as scheduling. In this work…
Orthogonal and Idempotent Transformations for Learning Deep Neural Networks
Jingdong Wang, Yajie Xing, Kexin Zhang +1
Identity transformations, used as skip-connections in residual networks, directly connect convolutional layers close to the input and those close to the output in deep neural netwo…