59 citations · 106 across the 27 of their papers we have counts for
27 papers
Ancient-Bench: A Comprehensive Multi-millennial, Multi-medium, and Multi-script Benchmark for Ancient Chinese Artifact Text Recognition
Hiuyi Cheng, Nuo Xu, Yuyi Zhang +9
Ancient Chinese artifact text recognition is fundamental to heritage digitization, and benchmarks for ancient texts are essential for evaluating current model capabilities. However…
TongGuOCR: A Layout-Aware and Token-Augmented OCR MLLM for Chinese Historical Documents
Zhongheng Zhou, Yi Sun, Huiguo He +6
Chinese historical documents preserve valuable cultural heritage, but many collections remain accessible only as scanned page images, preventing full-text retrieval, collation, and…
DiffMath: Symbol- and Graph-Aware Latent Diffusion Transformer for Handwritten Mathematical Expression Generation
Wei Pan, Xuhan Zheng, Yilin Shi +5
Handwritten Mathematical Expression Generation (HMEG) is challenging due to the complex two-dimensional layouts and long-range structural dependencies of mathematical expressions.…
Reinforcement Learning with Robust Rubric Rewards
Ya-Qi Yu, Hao Wang, Fangyu Hong +15
While Reinforcement Learning with Verifiable Rewards (RLVR) is effective for deterministically checkable tasks, many vision-language tasks are partially verifiable, demanding multi…
Visual Preference Optimization with Rubric Rewards
Ya-Qi Yu, Fangyu Hong, Xiangyang Qu +15
The effectiveness of Direct Preference Optimization (DPO) depends on preference data that reflect the quality differences that matter in multimodal tasks. Existing pipelines often…
URaG: Unified Retrieval and Generation in Multimodal LLMs for Efficient Long Document Understanding
Yongxin Shi, Jiapeng Wang, Zeyu Shan +3
Recent multimodal large language models (MLLMs) still struggle with long document understanding due to two fundamental challenges: information interference from abundant irrelevant…
OCR-Reasoning Benchmark: Unveiling the True Capabilities of MLLMs in Complex Text-Rich Image Reasoning
Mingxin Huang, Yongxin Shi, Dezhi Peng +3
Recent advancements in multimodal slow-thinking systems have demonstrated remarkable performance across various visual reasoning tasks. However, their capabilities in text-rich ima…
RedundancyLens: Revealing and Exploiting Visual Token Processing Redundancy for Efficient Decoder-Only MLLMs
Hongliang Li, Jiaxin Zhang, Wenhui Liao +3
Current Multimodal Large Language Model (MLLM) architectures face a critical tradeoff between performance and efficiency: decoder-only architectures achieve higher performance but…
Predicting the Original Appearance of Damaged Historical Documents
Zhenhua Yang, Dezhi Peng, Yongxin Shi +3
Historical documents encompass a wealth of cultural treasures but suffer from severe damages including character missing, paper damage, and ink erosion over time. However, existing…
TongGu: Mastering Classical Chinese Understanding with Knowledge-Grounded Large Language Models
Jiahuan Cao, Dezhi Peng, Peirong Zhang +4
Classical Chinese is a gateway to the rich heritage and wisdom of ancient China, yet its complexities pose formidable comprehension barriers for most modern people without speciali…
CBench: A Comprehensive Classical Chinese Understanding Benchmark for Large Language Models
Jiahuan Cao, Yongxin Shi, Dezhi Peng +2
Classical Chinese Understanding (CCU) holds significant value in preserving and exploration of the outstanding traditional Chinese culture. Recently, researchers have attempted to…
DocRes: A Generalist Model Toward Unifying Document Image Restoration Tasks
Jiaxin Zhang, Dezhi Peng, Chongyu Liu +2
Document image restoration is a crucial aspect of Document AI systems, as the quality of document images significantly influences the overall performance. Prevailing methods addres…
HierCode: A Lightweight Hierarchical Codebook for Zero-shot Chinese Text Recognition
Yuyi Zhang, Yuanzhi Zhu, Dezhi Peng +5
Text recognition, especially for complex scripts like Chinese, faces unique challenges due to its intricate character structures and vast vocabulary. Traditional one-hot encoding m…
SwinTextSpotter v2: Towards Better Synergy for Scene Text Spotting
Mingxin Huang, Dezhi Peng, Hongliang Li +6
End-to-end scene text spotting, which aims to read the text in natural images, has garnered significant attention in recent years. However, recent state-of-the-art methods usually…
FontDiffuser: One-Shot Font Generation via Denoising Diffusion with Multi-Scale Content Aggregation and Style Contrastive Learning
Zhenhua Yang, Dezhi Peng, Yuxin Kong +3
Automatic font generation is an imitation task, which aims to create a font library that mimics the style of reference images while preserving the content from source images. Altho…
UPOCR: Towards Unified Pixel-Level OCR Interface
Dezhi Peng, Zhenhua Yang, Jiaxin Zhang +5
Existing optical character recognition (OCR) methods rely on task-specific designs with divergent paradigms, architectures, and training strategies, which significantly increases t…
Exploring OCR Capabilities of GPT-4V(ision) : A Quantitative and In-depth Evaluation
Yongxin Shi, Dezhi Peng, Wenhui Liao +5
This paper presents a comprehensive evaluation of the Optical Character Recognition (OCR) capabilities of the recently released GPT-4V(ision), a Large Multimodal Model (LMM). We as…
ESTextSpotter: Towards Better Scene Text Spotting with Explicit Synergy in Transformer
Mingxin Huang, Jiaxin Zhang, Dezhi Peng +5
In recent years, end-to-end scene text spotting approaches are evolving to the Transformer-based framework. While previous studies have shown the crucial importance of the intrinsi…
Revisiting Scene Text Recognition: A Data Perspective
Qing Jiang, Jiapeng Wang, Dezhi Peng +2
This paper aims to re-assess scene text recognition (STR) from a data-oriented perspective. We begin by revisiting the six commonly used benchmarks in STR and observe a trend of pe…
ViTEraser: Harnessing the Power of Vision Transformers for Scene Text Removal with SegMIM Pretraining
Dezhi Peng, Chongyu Liu, Yuliang Liu +1
Scene text removal (STR) aims at replacing text strokes in natural scenes with visually coherent backgrounds. Recent STR approaches rely on iterative refinements or explicit text m…
SPTS v2: Single-Point Scene Text Spotting
Yuliang Liu, Jiaxin Zhang, Dezhi Peng +8
End-to-end scene text spotting has made significant progress due to its intrinsic synergy between text detection and recognition. Previous methods commonly regard manual annotation…
PageNet: Towards End-to-End Weakly Supervised Page-Level Handwritten Chinese Text Recognition
Dezhi Peng, Lianwen Jin, Yuliang Liu +2
Handwritten Chinese text recognition (HCTR) has been an active research topic for decades. However, most previous studies solely focus on the recognition of cropped text line image…
Recognition of Handwritten Chinese Text by Segmentation: A Segment-annotation-free Approach
Dezhi Peng, Lianwen Jin, Weihong Ma +4
Online and offline handwritten Chinese text recognition (HTCR) has been studied for decades. Early methods adopted oversegmentation-based strategies but suffered from low speed, in…
SLOGAN: Handwriting Style Synthesis for Arbitrary-Length and Out-of-Vocabulary Text
Canjie Luo, Yuanzhi Zhu, Lianwen Jin +2
Large amounts of labeled data are urgently required for the training of robust text recognizers. However, collecting handwriting data of diverse styles, along with an immense lexic…