Author name cluster

Ling Guo

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

5 papers

2 author rows

JBHI Journal 2026 Journal Article

HDPL: Hypergraph-based Dynamic Prompting Learning for Incomplete Multimodal Medical Learning

Xiaomin Zhou
Guoheng Huang
Qin Zhao
Jianbin He
Xiaochen Yuan
Ming Li
Chi-Man Pun
Ling Guo

Multimodal learning has garnered significant attention in the medical field due to its ability to provide a more comprehensive perspective utilizing various types of data, that aids in making more accurate decisions. However, the complexity of medical data, coupled with missing modalities, severely hinders predictive accuracy. Existing methods for multimodal learning with missing modalities still face considerable challenges. For instance, approaches that construct multimodal shared feature spaces often result in high computational costs, while methods that infer missing modalities based on complete ones may overly rely on the complete modalities, potentially skewing results. Pre-trained transformer methods address these issues but still have limitations, such as it can only process one missing modality at testing-stage. This is partly because structured data, unlike sequential data, lacks inherent minimum semantic units or natural order. Additionally, the positional encodings generated by this type of methods may introduce information interference when applied to structured data, leading to poor alignment with sequential data during modality fusion in transformer models. To tackle these challenges, we introduce HDPL: Hypergraph-based Dynamic Prompt Learning for Incomplete Multimodal Medical Learning, comprising three modules. The High-Order Hypergraph Embedding module can identify the minimal semantic units within structured data and utilizes hypergraph structures to extract high-dimensional features from clinical data. The Multimodal Medical Data Integrator module closes the distance of the embedding vectors corresponding in the shared space of modality-features, facilitating the integration of modalities in transformer. The Dynamic Network Structure Optimization module is a dynamic learning network by dynamically change the width and depth of network, improving the overall performance of the model, and it alleviates the shortcomings caused by incomplete modality to some extent. Through comprehensive experimentation, we demonstrate the efficiency and robustness of our model in dealing missing modalities and reducing training-burdens. Our code and dataset are available at https://github.com/colorful823/HDPL.

Details DOI

EAAI Journal 2026 Journal Article

Overcoming sparsity in micro ribonucleic acid–messenger ribonucleic acid regulatory networks: A graph sample and aggregate framework with learnable virtual nodes

Zhiping Sun
Ling Guo
Jia Mu
Tao Qu
Yue Pan

The mutual regulation between micro ribonucleic acid (miRNA) and messenger ribonucleic acid (mRNA) plays a critical role in tumorigenesis. However, the inherent sparsity of miRNA–mRNA regulatory networks is still a major challenge for computational prediction. This study proposes a Graph Sample and Aggregate (GraphSAGE) encoder–decoder architecture with a learnable virtual node mechanism. It is a novel graph neural network framework that improves global information flow and mitigates graph sparsity, termed Learnable Virtual node-enhanced GraphSAGE (LV-SAGE). The model is trained on high-confidence interactions derived from the intersection of four complementary differential expression analysis methods, as well as validated miRNA–mRNA pairs from comprehensive databases. Based on breast cancer and lung adenocarcinoma data, LV-SAGE achieves strong predictive performance with average Areas Under the Curve (AUC) of 97. 95% and 98. 08%, outperforming several existing methods. Systematic analysis further indicates that an appropriate number of virtual nodes with moderate connection ratios provides a practical balance between accuracy and efficiency. This framework provides a scalable and interpretable artificial intelligence approach for inferring sparse biological regulatory networks.

Details DOI

EAAI Journal 2025 Journal Article

Multi-order feature learning with global–local attention for real-time X-ray security inspection

Ling Guo
Yangbin Xu
Shouhong Chen

X-ray security inspection in high-throughput environments such as airports faces fundamental challenges stemming from the intrinsic complexity of baggage scanning. These challenges include feature entanglement from overlapping objects, visual distortions due to fixed imaging angles, inconsistent pseudo-color representations caused by varying X-ray absorption, and the difficulty of detecting prohibited items across multiple scales. To tackle these challenges, we introduce Multi-Order Gated Network (MoGNet), a lightweight transformer-based architecture that integrates three core innovations. First, a Multi-order Gated Aggregation Block designed for efficient multi-order feature extraction. Second, a Global–Local Self-Attention mechanism that enhances differentiation between foreground and background elements. Finally, a DynamicFusion module for adaptive integration of multi-scale features. Comprehensive evaluations on five challenging datasets establish our proposed framework, MoGNet, as the new state-of-the-art (SOTA). The model demonstrates superior performance, achieving mean Average Precision (mAP) scores at a 50% Intersection over Union (IoU) threshold (mAP50) of 75. 4%, 91. 9%, 91. 2%, 96. 6%, and 80. 0%, respectively. This high accuracy is maintained while operating at an efficient 64. 1 Frames Per Second (FPS). These comprehensive experimental results demonstrate its remarkable capability to optimize the balance between computational efficiency and detection accuracy, establishing it as a preferable solution for real-time contraband detection in practical security screening.

Details DOI

AAAI Conference 2021 Conference Paper

Frivolous Units: Wider Networks Are Not Really That Wide

Stephen Casper
Xavier Boix
Vanessa D'Amario
Ling Guo
Martin Schrimpf
Kasper Vinken
Gabriel Kreiman

A remarkable characteristic of overparameterized deep neural networks (DNNs) is that their accuracy does not degrade when the network width is increased. Recent evidence suggests that developing compressible representations allows the complexity of large networks to be adjusted for the learning task at hand. However, these representations are poorly understood. A promising strand of research inspired from biology involves studying representations at the unit level as it offers a more granular interpretation of the neural mechanisms. In order to better understand what facilitates increases in width without decreases in accuracy, we ask: Are there mechanisms at the unit level by which networks control their effective complexity? If so, how do these depend on the architecture, dataset, and hyperparameters? We identify two distinct types of “frivolous” units that proliferate when the network’s width increases: prunable units which can be dropped out of the network without significant change to the output and redundant units whose activities can be expressed as a linear combination of others. These units imply complexity constraints as the function the network computes could be expressed without them. We also identify how the development of these units can be influenced by architecture and a number of training factors. Together, these results help to explain why the accuracy of DNNs does not degrade when width is increased and highlight the importance of frivolous units toward understanding implicit regularization in DNNs.

PDF Details

IROS Conference 2010 Conference Paper

Design and calibration of single-camera catadioptric omnistereo system for miniature aerial vehicles (MAVs)

Ling Guo
Igor Labutov
Jizhong Xiao

Stereo system plays an important role in the navigation of MAVs. In this paper, we design a single-camera catadioptric omnistereo system for MAV, which consists of one hyperboloidal mirror, one hyperboloidal-planar combined mirror, and one conventional camera. System parameters are optimized based on the analysis of constraints and each parameter's influence on performance. Projective model of this system is derived, which provides a foundation for sphere-based calibration algorithm. It calibrates not only the conventional camera parameters, but also the mirror parameters. We also prove that a minimum of two spheres are needed to calibrate the seven parameters.

Details