arXiv subjects

Chih-Yu Wang

Publications and source records attributed to Chih-Yu Wang.

9 recordsLinked to original sources

arXiv2025

THeGAU: Type-Aware Heterogeneous Graph Autoencoder and Augmentation

Heterogeneous Graph Neural Networks (HGNNs) are effective for modeling Heterogeneous Information Networks (HINs), which encode complex multi-typed entities and relations. However, HGNNs often suffer from type information loss and structural noise, limiting their representational fidelity and generalization. We propose THeGAU, a model-agnostic framework that combines a type-aware graph autoencoder with guided graph augmentation to improve node classification. THeGAU reconstructs schema-valid edges as an auxiliary task to preserve node-type semantics and introduces a decoder-driven augmentation mechanism to selectively refine noisy structures. This joint design enhances robustness, accuracy, and efficiency while significantly reducing computational overhead. Extensive experiments on three benchmark HIN datasets (IMDB, ACM, and DBLP) demonstrate that THeGAU consistently outperforms existing HGNN methods, achieving state-of-the-art performance across multiple backbones.

cs.LG↗

arXiv2025

Decoherence-Aware Entangling and Swapping Strategy Optimization for Entanglement Routing in Quantum Networks

Quantum teleportation enables high-security communications through end-to-end quantum entangled pairs. End-to-end entangled pairs are created by using swapping processes to consume short entangled pairs and generate long pairs. However, due to environmental interference, entangled pairs decohere over time, resulting in low fidelity. Thus, generating entangled pairs at the right time is crucial. Moreover, the swapping process also causes additional fidelity loss. To this end, this paper presents a short time slot protocol, where a time slot can only accommodate a process. It has a more flexible arrangement of entangling and swapping processes than the traditional long time slot protocol. It raises a new optimization problem TETRIS for finding strategies of entangling and swapping for each request to maximize the fidelity sum of all accepted requests. To solve the TETRIS, we design two novel algorithms with different optimization techniques. Finally, the simulation results manifest that our algorithms can outperform the existing methods by up to 60 ~ 78% in general, and by 20 ~ 75% even under low entangling probabilities.

quant-ph↗

arXiv2025

Gen-n-Val: Agentic Image Data Generation and Validation

The data scarcity, label noise, and long-tailed category imbalance remain important and unresolved challenges in many computer vision tasks, such as object detection and instance segmentation, especially on large-vocabulary benchmarks like LVIS, where most categories appear in only a few images. Current synthetic data generation methods still suffer from multiple objects per mask, inaccurate segmentation, incorrect category labels, and other issues, limiting their effectiveness. To address these issues, we introduce Gen-n-Val, a novel agentic data generation framework that leverages Layer Diffusion (LD), a Large Language Model (LLM), and a Vision Large Language Model (VLLM) to produce high-quality and diverse instance masks and images for object detection and instance segmentation. Gen-n-Val consists of two agents: (1) the LD prompt agent, an LLM, optimizes rompts to encourage LD to generate high-quality foreground single-object images and corresponding segmentation masks; and (2) the data validation agent, a VLLM, filters out low-quality synthetic instance images. The system prompts for both agents are optimized by TextGrad. Compared to state-of-the-art synthetic data approaches like MosaicFusion, our approach reduces invalid synthetic data from 50% to 7% and improves performance by 7.6% on rare classes in LVIS instance segmentation with Mask R-CNN, and by 3.6% mAP on rare classes in COCO instance segmentation with YOLOv9c and YOLO11m. Furthermore, Gen-n-Val shows significant improvements (7.1% mAP) over YOLO-Worldv2-M in open-vocabulary object detection benchmarks with YOLO11m. Moreover, Gen-n-Val has scalability in model capacity and dataset size. The code is available at https://github.com/aiiu-lab/Gen-n-Val.

cs.CV↗

arXiv2025

Swapped Logit Distillation via Bi-level Teacher Alignment

Knowledge distillation (KD) compresses the network capacity by transferring knowledge from a large (teacher) network to a smaller one (student). It has been mainstream that the teacher directly transfers knowledge to the student with its original distribution, which can possibly lead to incorrect predictions. In this article, we propose a logit-based distillation via swapped logit processing, namely Swapped Logit Distillation (SLD). SLD is proposed under two assumptions: (1) the wrong prediction occurs when the prediction label confidence is not the maximum; (2) the "natural" limit of probability remains uncertain as the best value addition to the target cannot be determined. To address these issues, we propose a swapped logit processing scheme. Through this approach, we find that the swap method can be effectively extended to teacher and student outputs, transforming into two teachers. We further introduce loss scheduling to boost the performance of two teachers' alignment. Extensive experiments on image classification tasks demonstrate that SLD consistently performs best among previous state-of-the-art methods.

cs.LG↗

arXiv2024

StyleDiT: A Unified Framework for Diverse Child and Partner Faces Synthesis with Style Latent Diffusion Transformer

Kinship face synthesis is a challenging problem due to the scarcity and low quality of the available kinship data. Existing methods often struggle to generate descendants with both high diversity and fidelity while precisely controlling facial attributes such as age and gender. To address these issues, we propose the Style Latent Diffusion Transformer (StyleDiT), a novel framework that integrates the strengths of StyleGAN with the diffusion model to generate high-quality and diverse kinship faces. In this framework, the rich facial priors of StyleGAN enable fine-grained attribute control, while our conditional diffusion model is used to sample a StyleGAN latent aligned with the kinship relationship of conditioning images by utilizing the advantage of modeling complex kinship relationship distribution. StyleGAN then handles latent decoding for final face generation. Additionally, we introduce the Relational Trait Guidance (RTG) mechanism, enabling independent control of influencing conditions, such as each parent's facial image. RTG also enables a fine-grained adjustment between the diversity and fidelity in synthesized faces. Furthermore, we extend the application to an unexplored domain: predicting a partner's facial images using a child's image and one parent's image within the same framework. Extensive experiments demonstrate that our StyleDiT outperforms existing methods by striking an excellent balance between generating diverse and high-fidelity kinship faces.

cs.CV↗

arXiv2024

SynHING: Synthetic Heterogeneous Information Network Generation for Graph Learning and Explanation

Graph Neural Networks (GNNs) excel in delineating graph structures in diverse domains, including community analysis and recommendation systems. As the interpretation of GNNs becomes increasingly important, the demand for robust baselines and expansive graph datasets is accentuated, particularly in the context of Heterogeneous Information Networks (HIN). Addressing this, we introduce SynHING, a novel framework for Synthetic Heterogeneous Information Network Generation aimed at enhancing graph learning and explanation. SynHING systematically identifies major motifs in a target HIN and employs a bottom-up generation process with intra-cluster and inter-cluster merge modules. This process, supplemented by post-pruning techniques, ensures the synthetic HIN closely mirrors the original graph's structural and statistical properties. Crucially, SynHING provides ground-truth motifs for evaluating GNN explainer models, setting a new standard for explainable, synthetic HIN generation and contributing to the advancement of interpretable machine learning in complex networks.

cs.LG↗

arXiv2023

A GAN Approach for Node Embedding in Heterogeneous Graphs Using Subgraph Sampling

Graph neural networks (GNNs) face significant challenges with class imbalance, leading to biased inference results. To address this issue in heterogeneous graphs, we propose a novel framework that combines Graph Neural Network (GNN) and Generative Adversarial Network (GAN) to enhance classification for underrepresented node classes. The framework incorporates an advanced edge generation and selection module, enabling the simultaneous creation of synthetic nodes and edges through adversarial learning. Unlike previous methods, which predominantly focus on homogeneous graphs due to the difficulty of representing heterogeneous graph structures in matrix form, this approach is specifically designed for heterogeneous data. Existing solutions often rely on pre-trained models to incorporate synthetic nodes, which can lead to optimization inconsistencies and mismatches in data representation. Our framework avoids these pitfalls by generating data that aligns closely with the inherent graph topology and attributes, ensuring a more cohesive integration. Evaluations on multiple real-world datasets demonstrate the method's superiority over baseline models, particularly in tasks focused on identifying minority node classes, with notable improvements in performance metrics such as F-score and AUC-PRC score. These findings highlight the potential of this approach for addressing critical challenges in the field.

cs.LG↗

arXiv2012

Chinese Restaurant Game - Part I: Theory of Learning with Negative Network Externality

In a social network, agents are intelligent and have the capability to make decisions to maximize their utilities. They can either make wise decisions by taking advantages of other agents' experiences through learning, or make decisions earlier to avoid competitions from huge crowds. Both these two effects, social learning and negative network externality, play important roles in the decision process of an agent. While there are existing works on either social learning or negative network externality, a general study on considering both these two contradictory effects is still limited. We find that the Chinese restaurant process, a popular random process, provides a well-defined structure to model the decision process of an agent under these two effects. By introducing the strategic behavior into the non-strategic Chinese restaurant process, in Part I of this two-part paper, we propose a new game, called Chinese Restaurant Game, to formulate the social learning problem with negative network externality. Through analyzing the proposed Chinese restaurant game, we derive the optimal strategy of each agent and provide a recursive method to achieve the optimal strategy. How social learning and negative network externality influence each other under various settings is also studied through simulations.

cs.SI↗

arXiv2011

Chinese Restaurant Game - Part II: Applications to Wireless Networking, Cloud Computing, and Online Social Networking

In Part I of this two-part paper [1], we proposed a new game, called Chinese restaurant game, to analyze the social learning problem with negative network externality. The best responses of agents in the Chinese restaurant game with imperfect signals are constructed through a recursive method, and the influence of both learning and network externality on the utilities of agents is studied. In Part II of this two-part paper, we illustrate three applications of Chinese restaurant game in wireless networking, cloud computing, and online social networking. For each application, we formulate the corresponding problem as a Chinese restaurant game and analyze how agents learn and make strategic decisions in the problem. The proposed method is compared with four common-sense methods in terms of agents' utilities and the overall system performance through simulations. We find that the proposed Chinese restaurant game theoretic approach indeed helps agents make better decisions and improves the overall system performance. Furthermore, agents with different decision orders have different advantages in terms of their utilities, which also verifies the conclusions drawn in Part I of this two-part paper.

cs.SI↗