Xiangping Wu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Xiangping Wu has a total of 55 co-authors. Here are the top co-authors by number of shared publications:
- Qingcai Chen
Harbin Institute of Technology Shenzhen
16 shared publications - Xin Liu
Pengcheng Laboratory, Shenzhen, China
6 shared publications - Buzhou Tang
School of Computer Science and Technology, Harbin Institute of Technology, Shenzhen, China
3 shared publications - Fan Yang
School of Information and Communication Engineering, University of Electronic Science and Technology of China, Chengdu 611731, China
3 shared publications - Heng Li
School of Computer Science and Technology Harbin Institute of Technology (Shenzhen) Shenzhen China
3 shared publications - Shuoran Jiang
Harbin Institute of Technology, Shenzhen, China
3 shared publications
Latest Publications
- 30 July 2026
- Published by Elsevier in Pattern Recognition Letters
- Vol. 208, 157-164
FSE: Continual learning for named entity recognition by fast-slow experts
- 30 July 2026
- Published by Elsevier in Pattern Recognition Letters
- Vol. 208, 157-164
FSE: Continual learning for named entity recognition by fast-slow experts
- 26 November 2025
- Published by Elsevier in Applied Soft Computing
- Vol. 187
Compression of fully-connected layers based on parameter sharing for large-scale category image classification
- 26 November 2025
- Published by Elsevier in Applied Soft Computing
- Vol. 187
Compression of fully-connected layers based on parameter sharing for large-scale category image classification
- Open Access
GeRe: Towards Efficient Anti-Forgetting in Continual Learning of LLM via General Samples Replay
The continual learning capability of large language models (LLMs) is crucial for advancing artificial general intelligence. However, continual fine-tuning LLMs across various domains often suffers from catastrophic forgetting, characterized by: 1) significant forgetting of their general capabilities, and 2) sharp performance declines in previously learned tasks. To simultaneously address both issues in a simple yet stable manner, we propose General Sample Replay (GeRe), a framework that use usual pretraining texts for efficient anti-forgetting. Beyond revisiting the most prevalent replay-based practices under GeRe, we further leverage neural states to introduce a enhanced activation states constrained optimization method using threshold-based margin (TM) loss, which maintains activation state consistency during replay learning. We are the first to validate that a small, fixed set of pre-collected general replay samples is sufficient to resolve both concerns--retaining general capabilities while promoting overall performance across sequential tasks. Indeed, the former can inherently facilitate the latter. Through controlled experiments, we systematically compare TM with different replay strategies under the GeRe framework, including vanilla label fitting, logit imitation via KL divergence and feature imitation via L1/L2 losses. Results demonstrate that TM consistently improves performance and exhibits better robustness. Our work paves the way for efficient replay of LLMs for the future. Our code and data are available at https://github.com/Qznan/GeRe.
Open AccessGeRe: Towards Efficient Anti-Forgetting in Continual Learning of LLM via General Samples Replay
The continual learning capability of large language models (LLMs) is crucial for advancing artificial general intelligence. However, continual fine-tuning LLMs across various domains often suffers from catastrophic forgetting, characterized by: 1) significant forgetting of their general capabilities, and 2) sharp performance declines in previously learned tasks. To simultaneously address both issues in a simple yet stable manner, we propose General Sample Replay (GeRe), a framework that use usual pretraining texts for efficient anti-forgetting. Beyond revisiting the most prevalent replay-based practices under GeRe, we further leverage neural states to introduce a enhanced activation states constrained optimization method using threshold-based margin (TM) loss, which maintains activation state consistency during replay learning. We are the first to validate that a small, fixed set of pre-collected general replay samples is sufficient to resolve both concerns--retaining general capabilities while promoting overall performance across sequential tasks. Indeed, the former can inherently facilitate the latter. Through controlled experiments, we systematically compare TM with different replay strategies under the GeRe framework, including vanilla label fitting, logit imitation via KL divergence and feature imitation via L1/L2 losses. Results demonstrate that TM consistently improves performance and exhibits better robustness. Our work paves the way for efficient replay of LLMs for the future. Our code and data are available at https://github.com/Qznan/GeRe.
- Open Access
- 30 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 1293-1302
Local and Global Aware Document Image Enhancement with Residual Denoising Diffusion Model
- 30 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 1293-1302
Open AccessLocal and Global Aware Document Image Enhancement with Residual Denoising Diffusion Model
- 11 May 2025
- Published by Springer Nature in Optical Review
- p. 1-16
Adaptive cross-modal fusion for robust multi-modal object detection in infrared–visible imaging
- 11 May 2025
- Published by Springer Nature in Optical Review
- p. 1-16
Adaptive cross-modal fusion for robust multi-modal object detection in infrared–visible imaging
- Open Access
Document Image Rectification Bases on Self-Adaptive Multitask Fusion
Deformed document image rectification is essential for real-world document understanding tasks, such as layout analysis and text recognition. However, current multi-task methods -- such as background removal, 3D coordinate prediction, and text line segmentation -- often overlook the complementary features between tasks and their interactions. To address this gap, we propose a self-adaptive learnable multi-task fusion rectification network named SalmRec. This network incorporates an inter-task feature aggregation module that adaptively improves the perception of geometric distortions, enhances feature complementarity, and reduces negative interference. We also introduce a gating mechanism to balance features both within global tasks and between local tasks effectively. Experimental results on two English benchmarks (DIR300 and DocUNet) and one Chinese benchmark (DocReal) demonstrate that our method significantly improves rectification performance. Ablation studies further highlight the positive impact of different tasks on dewarping and the effectiveness of our proposed module.
Open AccessDocument Image Rectification Bases on Self-Adaptive Multitask Fusion
Deformed document image rectification is essential for real-world document understanding tasks, such as layout analysis and text recognition. However, current multi-task methods -- such as background removal, 3D coordinate prediction, and text line segmentation -- often overlook the complementary features between tasks and their interactions. To address this gap, we propose a self-adaptive learnable multi-task fusion rectification network named SalmRec. This network incorporates an inter-task feature aggregation module that adaptively improves the perception of geometric distortions, enhances feature complementarity, and reduces negative interference. We also introduce a gating mechanism to balance features both within global tasks and between local tasks effectively. Experimental results on two English benchmarks (DIR300 and DocUNet) and one Chinese benchmark (DocReal) demonstrate that our method significantly improves rectification performance. Ablation studies further highlight the positive impact of different tasks on dewarping and the effectiveness of our proposed module.
- Open Access
- 9 November 2024
- Published by Springer Nature in Complex & Intelligent Systems
- Vol. 11 (1) , 1-13
Travel route recommendation with a trajectory learning model
- 9 November 2024
- Published by Springer Nature in Complex & Intelligent Systems
- Vol. 11 (1) , 1-13
Open AccessTravel route recommendation with a trajectory learning model
- Open Access
- 17 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 86096-86105
Incentive Instruction for Event Relation Extraction in Low-Resource
Relation extraction is the foundation for constructing a Event Evolution Graph. However, existing methods for extracting relation are more focus on data sufficient. This paper proposes a method for extracting relation in low-resource domain. Firstly, a incentive instruction is designed for task. We incorporate incentive instruction into the original text, which contain factual descriptions of the relation between defective events and the relation that need to be predicted. Additionally, defective events are included in the incentive instruction, allowing the model to shift its focus from the contextual representation of individual event words to the overall representation of text units. By employing this approach, we transform the relation extraction task into a pre-training task. Specifically, the Bidirectional Encoder Representation from Transformers places relation prediction words in designated positions within the incentive instructions and generates labels for each prediction word. Finally, we obtain the optimal relation. Compared with other models, the proposed method improves the F1 scores of the three logical relations by 14.41% to 23.19%, 15.45% to 28.42%, and 15.33% to 27.02%, respectively.
- 17 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 86096-86105
Open AccessIncentive Instruction for Event Relation Extraction in Low-Resource
Relation extraction is the foundation for constructing a Event Evolution Graph. However, existing methods for extracting relation are more focus on data sufficient. This paper proposes a method for extracting relation in low-resource domain. Firstly, a incentive instruction is designed for task. We incorporate incentive instruction into the original text, which contain factual descriptions of the relation between defective events and the relation that need to be predicted. Additionally, defective events are included in the incentive instruction, allowing the model to shift its focus from the contextual representation of individual event words to the overall representation of text units. By employing this approach, we transform the relation extraction task into a pre-training task. Specifically, the Bidirectional Encoder Representation from Transformers places relation prediction words in designated positions within the incentive instructions and generates labels for each prediction word. Finally, we obtain the optimal relation. Compared with other models, the proposed method improves the F1 scores of the three logical relations by 14.41% to 23.19%, 15.45% to 28.42%, and 15.33% to 27.02%, respectively.
- 28 May 2024
- Published by SAGE Publications in Intelligent Data Analysis: An International Journal
- Vol. 28 (3) , 825-840
Generate custom travel magazine layouts
Among the problems of specifying the style and number of elements of a travel magazine, the problem of generating magazine layout by constraining text, and constraining graph layout remains a complex and unsolved problem. In this paper, we generate layouts of text satisfying constraints via GAN. Due to the complexity and variety of graph designs, we enhance the performance of the discriminator and the generator so that the layouts generated by the generator are more constrained. Add non-corresponding constraint text and real layout pairs to the discriminator to enhance the performance of the discriminator; then add a spatial attention mechanism to the layout encoder to extract the features of the layout and generate high-quality layouts. We demonstrate that the proposed method can generate high-quality layouts of text satisfying the constraints, and we validate the effectiveness of this method through user ratings.
- 28 May 2024
- Published by SAGE Publications in Intelligent Data Analysis: An International Journal
- Vol. 28 (3) , 825-840
Generate custom travel magazine layouts
Among the problems of specifying the style and number of elements of a travel magazine, the problem of generating magazine layout by constraining text, and constraining graph layout remains a complex and unsolved problem. In this paper, we generate layouts of text satisfying constraints via GAN. Due to the complexity and variety of graph designs, we enhance the performance of the discriminator and the generator so that the layouts generated by the generator are more constrained. Add non-corresponding constraint text and real layout pairs to the discriminator to enhance the performance of the discriminator; then add a spatial attention mechanism to the layout encoder to extract the features of the layout and generate high-quality layouts. We demonstrate that the proposed method can generate high-quality layouts of text satisfying the constraints, and we validate the effectiveness of this method through user ratings.