arXiv subjects

Yanchun Zhang

Publications and source records attributed to Yanchun Zhang.

16 recordsLinked to original sources

arXiv2026

PPU-Bench:Real World Benchmark for Personalized Partial Unlearning in Vision Language Models

Multimodal Large Language Models (MLLMs) may memorize sensitive cross-modal information during pretraining. However, existing MLLM unlearning benchmarks rely on synthetic knowledge injection or complete subject-level deletion, which fail to capture realistic, personalized deletion requests that require fine-grained factual control. In this paper, we introduce PPU-Bench, a real-world and fine-tuning-free benchmark for personalized partial unlearning in MLLMs. PPU-Bench contains 24K multimodal and unimodal samples derived from pre-existing knowledge of 500 public figures under three progressively challenging settings: Complete, Selective, and Personalized unlearning. The benchmark evaluates whether methods can remove target knowledge while preserving non-target facts, model utility, and cross-modal consistency. Extensive experiments show that Complete Unlearning often suppresses visual identity rather than factual knowledge, while Selective and Personalized Unlearning expose significant forget--retain trade-offs and challenges in intra-subject factual boundaries. Robustness analysis under cross-image and prompt-based attacks reveals distinct vulnerabilities across different unlearning settings. Motivated by these findings, we propose Boundary-Aware Optimization (BAO), which explicitly models intra-subject forget-retain boundaries. Experimental results on two representative methods demonstrate that BAO can effectively enforce intra-subject factual boundaries.

cs.CV↗

arXiv2025

ABG-NAS: Adaptive Bayesian Genetic Neural Architecture Search for Graph Representation Learning

Effective and efficient graph representation learning is essential for enabling critical downstream tasks, such as node classification, link prediction, and subgraph search. However, existing graph neural network (GNN) architectures often struggle to adapt to diverse and complex graph structures, limiting their ability to produce structure-aware and task-discriminative representations. To address this challenge, we propose ABG-NAS, a novel framework for automated graph neural network architecture search tailored for efficient graph representation learning. ABG-NAS encompasses three key components: a Comprehensive Architecture Search Space (CASS), an Adaptive Genetic Optimization Strategy (AGOS), and a Bayesian-Guided Tuning Module (BGTM). CASS systematically explores diverse propagation (P) and transformation (T) operations, enabling the discovery of GNN architectures capable of capturing intricate graph characteristics. AGOS dynamically balances exploration and exploitation, ensuring search efficiency and preserving solution diversity. BGTM further optimizes hyperparameters periodically, enhancing the scalability and robustness of the resulting architectures. Empirical evaluations on benchmark datasets (Cora, PubMed, Citeseer, and CoraFull) demonstrate that ABG-NAS consistently outperforms both manually designed GNNs and state-of-the-art neural architecture search (NAS) methods. These results highlight the potential of ABG-NAS to advance graph representation learning by providing scalable and adaptive solutions for diverse graph structures. Our code is publicly available at https://github.com/sserranw/ABG-NAS.

cs.LG↗

arXiv2024

MDU-Net: Multi-scale Densely Connected U-Net for biomedical image segmentation

Biomedical image segmentation plays a central role in quantitative analysis, clinical diagnosis, and medical intervention. In the light of the fully convolutional networks (FCN) and U-Net, deep convolutional networks (DNNs) have made significant contributions to biomedical image segmentation applications. In this paper, we propose three different multi-scale dense connections (MDC) for the encoder, the decoder of U-shaped architectures, and across them. Based on three dense connections, we propose a multi-scale densely connected U-Net (MDU-Net) for biomedical image segmentation. MDU-Net directly fuses the neighboring feature maps with different scales from both higher layers and lower layers to strengthen feature propagation in the current layer. Multi-scale dense connections, which contain shorter connections between layers close to the input and output, also make a much deeper U-Net possible. Besides, we introduce quantization to alleviate the potential overfitting in dense connections, and further improve the segmentation performance. We evaluate our proposed model on the MICCAI 2015 Gland Segmentation (GlaS) dataset. The three MDC improve U-Net performance by up to 1.8% on test A and 3.5% on test B in the MICCAI Gland dataset. Meanwhile, the MDU-Net with quantization obviously improves the segmentation performance of original U-Net.

cs.CV↗

arXiv2024

Exploring Advanced Methodologies in Security Evaluation for LLMs

Large Language Models (LLMs) represent an advanced evolution of earlier, simpler language models. They boast enhanced abilities to handle complex language patterns and generate coherent text, images, audios, and videos. Furthermore, they can be fine-tuned for specific tasks. This versatility has led to the proliferation and extensive use of numerous commercialized large models. However, the rapid expansion of LLMs has raised security and ethical concerns within the academic community. This emphasizes the need for ongoing research into security evaluation during their development and deployment. Over the past few years, a substantial body of research has been dedicated to the security evaluation of large-scale models. This article an in-depth review of the most recent advancements in this field, providing a comprehensive analysis of commonly used evaluation metrics, advanced evaluation frameworks, and the routine evaluation processes for LLMs. Furthermore, we also discuss the future directions for advancing the security evaluation of LLMs.

cs.CR↗

arXiv2021

Efficient and Optimal Algorithms for Tree Summarization with Weighted Terminologies

Data summarization that presents a small subset of a dataset to users has been widely applied in numerous applications and systems. Many datasets are coded with hierarchical terminologies, e.g., the international classification of Diseases-9, Medical Subject Heading, and Gene Ontology, to name a few. In this paper, we study the problem of selecting a diverse set of k elements to summarize an input dataset with hierarchical terminologies, and visualize the summary in an ontology structure. We propose an efficient greedy algorithm to solve the problem with (1-1/e) = 62% -approximation guarantee. Although this greedy solution achieves quality-guaranteed answers approximately but it is still not optimal. To tackle the problem optimally, we further develop a dynamic programming algorithm to obtain optimal answers for graph visualization of log-data using ontology terminologies called OVDO . The complexity and correctness of OVDO are theoretically analyzed. In addition, we propose a useful optimization technique of tree reduction to remove useless nodes with zero weights and shrink the tree into a smaller one, which ensures the efficiency acceleration of OVDO in many real-world applications. Extensive experimental results on real-world datasets show the effectiveness and efficiency of our proposed approximate and exact algorithms for tree data summarization.

cs.DB↗

arXiv2021

ObjectAug: Object-level Data Augmentation for Semantic Image Segmentation

Semantic image segmentation aims to obtain object labels with precise boundaries, which usually suffers from overfitting. Recently, various data augmentation strategies like regional dropout and mix strategies have been proposed to address the problem. These strategies have proved to be effective for guiding the model to attend on less discriminative parts. However, current strategies operate at the image level, and objects and the background are coupled. Thus, the boundaries are not well augmented due to the fixed semantic scenario. In this paper, we propose ObjectAug to perform object-level augmentation for semantic image segmentation. ObjectAug first decouples the image into individual objects and the background using the semantic labels. Next, each object is augmented individually with commonly used augmentation methods (e.g., scaling, shifting, and rotation). Then, the black area brought by object augmentation is further restored using image inpainting. Finally, the augmented objects and background are assembled as an augmented image. In this way, the boundaries can be fully explored in the various semantic scenarios. In addition, ObjectAug can support category-aware augmentation that gives various possibilities to objects in each category, and can be easily combined with existing image-level augmentation methods to further boost performance. Comprehensive experiments are conducted on both natural image and medical image datasets. Experiment results demonstrate that our ObjectAug can evidently improve segmentation performance.

cs.CV↗

arXiv2021

Pyramid U-Net for Retinal Vessel Segmentation

Retinal blood vessel can assist doctors in diagnosis of eye-related diseases such as diabetes and hypertension, and its segmentation is particularly important for automatic retinal image analysis. However, it is challenging to segment these vessels structures, especially the thin capillaries from the color retinal image due to low contrast and ambiguousness. In this paper, we propose pyramid U-Net for accurate retinal vessel segmentation. In pyramid U-Net, the proposed pyramid-scale aggregation blocks (PSABs) are employed in both the encoder and decoder to aggregate features at higher, current and lower levels. In this way, coarse-to-fine context information is shared and aggregated in each block thus to improve the location of capillaries. To further improve performance, two optimizations including pyramid inputs enhancement and deep pyramid supervision are applied to PSABs in the encoder and decoder, respectively. For PSABs in the encoder, scaled input images are added as extra inputs. While for PSABs in the decoder, scaled intermediate outputs are supervised by the scaled segmentation labels. Extensive evaluations show that our pyramid U-Net outperforms the current state-of-the-art methods on the public DRIVE and CHASE-DB1 datasets.

eess.IV↗

arXiv2020

THCluster: herb supplements categorization for precision traditional Chinese medicine

There has been a continuing demand for traditional and complementary medicine worldwide. A fundamental and important topic in Traditional Chinese Medicine (TCM) is to optimize the prescription and to detect herb regularities from TCM data. In this paper, we propose a novel clustering model to solve this general problem of herb categorization, a pivotal task of prescription optimization and herb regularities. The model utilizes Random Walks method, Bayesian rules and Expectation Maximization(EM) models to complete a clustering analysis effectively on a heterogeneous information network. We performed extensive experiments on the real-world datasets and compared our method with other algorithms and experts. Experimental results have demonstrated the effectiveness of the proposed model for discovering useful categorization of herbs and its potential clinical manifestations.

q-bio.QM↗

arXiv2020

Exploiting Review Neighbors for Contextualized Helpfulness Prediction

Helpfulness prediction techniques have been widely used to identify and recommend high-quality online reviews to customers. Currently, the vast majority of studies assume that a review's helpfulness is self-contained. In practice, however, customers hardly process reviews independently given the sequential nature. The perceived helpfulness of a review is likely to be affected by its sequential neighbors (i.e., context), which has been largely ignored. This paper proposes a new methodology to capture the missing interaction between reviews and their neighbors. The first end-to-end neural architecture is developed for neighbor-aware helpfulness prediction (NAP). For each review, NAP allows for three types of neighbor selection: its preceding, following, and surrounding neighbors. Four weighting schemes are designed to learn context clues from the selected neighbors. A review is then contextualized into the learned clues for neighbor-aware helpfulness prediction. NAP is evaluated on six domains of real-world online reviews against a series of state-of-the-art baselines. Extensive experiments confirm the effectiveness of NAP and the influence of sequential neighbors on a current reviews. Further hyperparameter analysis reveals three main findings. (1) On average, eight neighbors treated with uneven importance are engaged for context construction. (2) The benefit of neighbor-aware prediction mainly results from closer neighbors. (3) Equally considering up to five closest neighbors of a review can usually produce a weaker but tolerable prediction result.

cs.CL↗

arXiv2019

Online Social Media Recommendation over Streams

As one of the most popular services over online communities, the social recommendation has attracted increasing research efforts recently. Among all the recommendation tasks, an important one is social item recommendation over high speed social media streams. Existing streaming recommendation techniques are not effective for handling social users with diverse interests. Meanwhile, approaches for recommending items to a particular user are not efficient when applied to a huge number of users over high speed streams. In this paper, we propose a novel framework for the social recommendation over streaming environments. Specifically, we first propose a novel Bi-Layer Hidden Markov Model (BiHMM) that adaptively captures the behaviors of social users and their interactions with influential official accounts to predict their long-term and short-term interests. Then, we design a new probabilistic entity matching scheme for effectively identifying the relevance score of a streaming item to a user. Following that, we propose a novel indexing scheme called {\Tree} for improving the efficiency of our solution. Extensive experiments are conducted to prove the high performance of our approach in terms of the recommendation quality and time cost.

cs.IR↗

arXiv2018

A Study of Tourist Sequential Activity Patterns through Location Based Social Network (LBSN)

Sequential Pattern Mining is an important component in establishing patterns and mining trends of certain activities. Insights into tourist movement and activity patterns is deemed beneficial for the tourism sector in many ways, such as designing better travel packages for tourists, maximizing the tourist activity participation and meeting the tourist demands. This research proposes to adopt mobile social media data for effective capturing of tourist activity information in Singapore and utilizes advanced data mining techniques for extracting valuable insights into tourist behavior. The proposed methods and findings of the study have the potential to support tourism managers and policy makers in making better decisions in tourism destination management.

cs.CY↗

arXiv2018

Preserving Patient-centred Controls in Electronic Health Record Systems: A Reliance-based Model Implication

As a consequence of the huge advancement of the Electronic Health Record (EHR) in healthcare settings, the My Health Record (MHR) is introduced in Australia. However security and privacy of the MHR system have been encumbering the development of the system. Even though the MHR system is claimed as patient-cenred and patient-controlled, there are several instances where healthcare providers (other than the usual provider) and system operators who maintain the system can easily access the system and these unauthorised accesses can lead to a breach of the privacy of the patients. This is one of the main concerns of the consumers that affect the uptake of the system. In this paper, we propose a patient centred MHR framework which requests authorisation from the patient to access their sensitive health information. The proposed model increases the involvement and satisfaction of the patients in their healthcare and also suggests mobile security system to give an online permission to access the MHR system.

cs.CY↗

arXiv2016

Superiorized iteration based on proximal point method and its application to XCT image reconstruction

In this paper, we investigate how to determine a better perturbation for superiorized iteration. We propose to seek the perturbation by proximal point method. In our method, the direction and amount of perturbation are computed simultaneously. The convergence conditions are also discussed for bounded perterbation resilence iteration. Numerical experiments on simulated XCT projection data show that the proposed method improves the convergence rate and the image quality.

math.NA↗

arXiv2016

Supervised Anomaly Detection in Uncertain Pseudoperiodic Data Streams

Uncertain data streams have been widely generated in many Web applications. The uncertainty in data streams makes anomaly detection from sensor data streams far more challenging. In this paper, we present a novel framework that supports anomaly detection in uncertain data streams. The proposed framework adopts an efficient uncertainty pre-processing procedure to identify and eliminate uncertainties in data streams. Based on the corrected data streams, we develop effective period pattern recognition and feature extraction techniques to improve the computational efficiency. We use classification methods for anomaly detection in the corrected data stream. We also empirically show that the proposed approach shows a high accuracy of anomaly detection on a number of real datasets.

cs.AI↗

arXiv2015

Refining Adverse Drug Reactions using Association Rule Mining for Electronic Healthcare Data

Side effects of prescribed medications are a common occurrence. Electronic healthcare databases present the opportunity to identify new side effects efficiently but currently the methods are limited due to confounding (i.e. when an association between two variables is identified due to them both being associated to a third variable). In this paper we propose a proof of concept method that learns common associations and uses this knowledge to automatically refine side effect signals (i.e. exposure-outcome associations) by removing instances of the exposure-outcome associations that are caused by confounding. This leaves the signal instances that are most likely to correspond to true side effect occurrences. We then calculate a novel measure termed the confounding-adjusted risk value, a more accurate absolute risk value of a patient experiencing the outcome within 60 days of the exposure. Tentative results suggest that the method works. For the four signals (i.e. exposure-outcome associations) investigated we are able to correctly filter the majority of exposure-outcome instances that were unlikely to correspond to true side effects. The method is likely to improve when tuning the association rule mining parameters for specific health outcomes. This paper shows that it may be possible to filter signals at a patient level based on association rules learned from considering patients' medical histories. However, additional work is required to develop a way to automate the tuning of the method's parameters.

cs.DB↗

arXiv2014

Efficient Truss Maintenance in Evolving Networks

Truss was proposed to study social network data represented by graphs. A k-truss of a graph is a cohesive subgraph, in which each edge is contained in at least k-2 triangles within the subgraph. While truss has been demonstrated as superior to model the close relationship in social networks and efficient algorithms for finding trusses have been extensively studied, very little attention has been paid to truss maintenance. However, most social networks are evolving networks. It may be infeasible to recompute trusses from scratch from time to time in order to find the up-to-date $k$-trusses in the evolving networks. In this paper, we discuss how to maintain trusses in a graph with dynamic updates. We first discuss a set of properties on maintaining trusses, then propose algorithms on maintaining trusses on edge deletions and insertions, finally, we discuss truss index maintenance. We test the proposed techniques on real datasets. The experiment results show the promise of our work.

cs.DB↗