arXiv subjects

Ying Zhang

Publications and source records attributed to Ying Zhang.

At least 199 records · Page 11Linked to original sources

arXiv2025

Transforming Credit Risk Analysis: A Time-Series-Driven ResE-BiLSTM Framework for Post-Loan Default Detection

Prediction of post-loan default is an important task in credit risk management, and can be addressed by detection of financial anomalies using machine learning. This study introduces a ResE-BiLSTM model, using a sliding window technique, and is evaluated on 44 independent cohorts from the extensive Freddie Mac US mortgage dataset, to improve prediction performance. The ResE-BiLSTM is compared with five baseline models: Long Short-Term Memory (LSTM), BiLSTM, Gated Recurrent Units (GRU), Convolutional Neural Networks (CNN), and Recurrent Neural Networks (RNN), across multiple metrics, including Accuracy, Precision, Recall, F1, and AUC. An ablation study was conducted to evaluate the contribution of individual components in the ResE-BiLSTM architecture. Additionally, SHAP analysis was employed to interpret the underlying features the model relied upon for its predictions. Experimental results demonstrate that ResE-BiLSTM achieves superior predictive performance compared to baseline models, underscoring its practical value and applicability in real-world scenarios.

cs.LG↗

arXiv2025

Investigating Crossing Perception in 3D Graph Visualisation

Human perception of graph drawings is influenced by a variety of impact factors for which quality measures are used as a proxy indicator. The investigation of those impact factors and their effects is important to evaluate and improve quality measures and drawing algorithms. The number of edge crossings in a 2D graph drawing has long been a main quality measure for drawing evaluation. The use of stereoscopic 3D graph visualisations has gained attraction over the last years, and results from several studies indicate that they can improve analysis efficiency for a range of analysis scenarios. While edge crossings can also occur in 3D, there are edge configurations in space that are not crossings but might be perceived as such from a specific viewpoint. Such configurations create crossings when projected on the corresponding 2D image plane and could impact readability similar to 2D crossings. In 3D drawings, the additional depth aspect and the subsequent impact factors of edge distance and relative edge direction in space might further influence the importance of those configurations for readability. We investigate the impact of such factors in an empirical study and report on findings of difference between major factor categories.

cs.GR↗

arXiv2025

Mesh based segmentation for automated margin line generation on incisors receiving crown treatment

Dental crowns are essential dental treatments for restoring damaged or missing teeth of patients. Recent design approaches of dental crowns are carried out using commercial dental design software. Once a scan of a preparation is uploaded to the software, a dental technician needs to manually define a precise margin line on the preparation surface, which constitutes a non-repeatable and inconsistent procedure. This work proposes a new framework to determine margin lines automatically and accurately using deep learning. A dataset of incisor teeth was provided by a collaborating dental laboratory to train a deep learning segmentation model. A mesh-based neural network was modified by changing its input channels and used to segment the prepared tooth into two regions such that the margin line is contained within the boundary faces separating the two regions. Next, k-fold cross-validation was used to train 5 models, and a voting classifier technique was used to combine their results to enhance the segmentation. After that, boundary smoothing and optimization using the graph cut method were applied to refine the segmentation results. Then, boundary faces separating the two regions were selected to represent the margin line faces. A spline was approximated to best fit the centers of the boundary faces to predict the margin line. Our results show that an ensemble model combined with maximum probability predicted the highest number of successful test cases (7 out of 13) based on a maximum distance threshold of 200 m (representing human error) between the predicted and ground truth point clouds. It was also demonstrated that the better the quality of the preparation, the smaller the divergence between the predicted and ground truth margin lines (Spearman's rank correlation coefficient of -0.683). We provide the train and test datasets for the community.

cs.CE↗

arXiv2025

WOCD: A Semi-Supervised Method for Overlapping Community Detection Using Weak Cliques

Overlapping community detection (OCD) is a fundamental graph data analysis task for extracting graph patterns. Traditional OCD methods can be broadly divided into node clustering and link clustering approaches, both of which rely solely on link information to identify overlapping communities. In recent years, deep learning-based methods have made significant advancements for this task. However, existing GNN-based approaches often face difficulties in effectively integrating link, attribute, and prior information, along with challenges like limited receptive fields and over-smoothing, which hinder their performance on complex overlapping community detection. In this paper, we propose a Weak-clique based Overlapping Community Detection method, namely WOCD, which incorporates prior information and optimizes the use of link information to improve detection accuracy. Specifically, we introduce pseudo-labels within a semi-supervised framework to strengthen the generalization ability, making WOCD more versatile. Furthermore, we initialize pseudo-labels using weak cliques to fully leverage link and prior information, leading to better detection accuracy. Additionally, we employ a single-layer Graph Transformer combined with GNN, which achieves significant performance improvements while maintaining efficiency. We evaluate WOCD on eight real-world attributed datasets, and the results demonstrate that it outperforms the state-of-the-art semi-supervised OCD method by a significant margin in terms of accuracy.

cs.SI↗

arXiv2025

Dark Side of Modalities: Reinforced Multimodal Distillation for Multimodal Knowledge Graph Reasoning

The multimodal knowledge graph reasoning (MKGR) task aims to predict the missing facts in the incomplete MKGs by leveraging auxiliary images and descriptions of entities. Existing approaches are trained with single-target objectives, which neglect the probabilistic correlations of entity labels, especially in non-target entities. Moreover, previous studies incorporate all modalities statically or adaptively, overlooking the negative impacts of irrelevant or misleading information in the incompetent modalities. To address these issues, we introduce a novel Reinforced Multimodal Distillation framework, exploiting the Dark Side of Modalities (DSoM) from two perspectives: (1) Dark knowledge from non-target entities: We propose to train a unimodal KGR model through logit distillation to mimic the multimodal soft labels provided by pre-trained multimodal teacher models. The multimodal soft labels could provide rich supervision signals with subtle correlations among both target and non-target entities from multiple perspectives. We further decouple logits into neighbor entities and non-neighbor entities to divide into two types of correlations. (2) Dark side in unhelpful modalities: To exclude the adverse effects of unhelpful modalities, we introduce a reinforced teacher combination mechanism that dynamically selects the optimal set of multimodal teachers for each triple. The agent is trained to maximize the rewards, which are only assigned to the beneficial multimodal combination strategies for the student model. Comprehensive experiments demonstrate the effectiveness of DSoM framework on 5 MKGR datasets. Codes are available at github.com/OreOZhao/DSoM.

cs.MM↗

arXiv2025

Regularizing Subspace Redundancy of Low-Rank Adaptation

Low-Rank Adaptation (LoRA) and its variants have delivered strong capability in Parameter-Efficient Transfer Learning (PETL) by minimizing trainable parameters and benefiting from reparameterization. However, their projection matrices remain unrestricted during training, causing high representation redundancy and diminishing the effectiveness of feature adaptation in the resulting subspaces. While existing methods mitigate this by manually adjusting the rank or implicitly applying channel-wise masks, they lack flexibility and generalize poorly across various datasets and architectures. Hence, we propose ReSoRA, a method that explicitly models redundancy between mapping subspaces and adaptively Regularizes Subspace redundancy of Low-Rank Adaptation. Specifically, it theoretically decomposes the low-rank submatrices into multiple equivalent subspaces and systematically applies de-redundancy constraints to the feature distributions across different projections. Extensive experiments validate that our proposed method consistently facilitates existing state-of-the-art PETL methods across various backbones and datasets in vision-language retrieval and standard visual classification benchmarks. Besides, as a training supervision, ReSoRA can be seamlessly integrated into existing approaches in a plug-and-play manner, with no additional inference costs. Code is publicly available at: https://github.com/Lucenova/ReSoRA.

cs.CV↗

arXiv2025

SAM2-Aug: Prior knowledge-based Augmentation for Target Volume Auto-Segmentation in Adaptive Radiation Therapy Using Segment Anything Model 2

Purpose: Accurate tumor segmentation is vital for adaptive radiation therapy (ART) but remains time-consuming and user-dependent. Segment Anything Model 2 (SAM2) shows promise for prompt-based segmentation but struggles with tumor accuracy. We propose prior knowledge-based augmentation strategies to enhance SAM2 for ART. Methods: Two strategies were introduced to improve SAM2: (1) using prior MR images and annotations as contextual inputs, and (2) improving prompt robustness via random bounding box expansion and mask erosion/dilation. The resulting model, SAM2-Aug, was fine-tuned and tested on the One-Seq-Liver dataset (115 MRIs from 31 liver cancer patients), and evaluated without retraining on Mix-Seq-Abdomen (88 MRIs, 28 patients) and Mix-Seq-Brain (86 MRIs, 37 patients). Results: SAM2-Aug outperformed convolutional, transformer-based, and prompt-driven models across all datasets, achieving Dice scores of 0.86(liver), 0.89(abdomen), and 0.90(brain). It demonstrated strong generalization across tumor types and imaging sequences, with improved performance in boundary-sensitive metrics. Conclusions: Incorporating prior images and enhancing prompt diversity significantly boosts segmentation accuracy and generalizability. SAM2-Aug offers a robust, efficient solution for tumor segmentation in ART. Code and models will be released at https://github.com/apple1986/SAM2-Aug.

eess.IV↗

arXiv2025

CP-OFDM Achieves the Lowest Average Ranging Sidelobe Under QAM/PSK Constellations

This paper aims to answer a fundamental question in the area of Integrated Sensing and Communications (ISAC): What is the optimal communication-centric ISAC waveform for ranging? Towards that end, we first established a generic framework to analyze the sensing performance of communication-centric ISAC waveforms built upon orthonormal signaling bases and random data symbols. Then, we evaluated their ranging performance by adopting both the periodic and aperiodic auto-correlation functions (P-ACF and A-ACF), and defined the expectation of the integrated sidelobe level (EISL) as a sensing performance metric. On top of that, we proved that among all communication waveforms with cyclic prefix (CP), the orthogonal frequency division multiplexing (OFDM) modulation is the only globally optimal waveform that achieves the lowest ranging sidelobe for quadrature amplitude modulation (QAM) and phase shift keying (PSK) constellations, in terms of both the EISL and the sidelobe level at each individual lag of the P-ACF. As a step forward, we proved that among all communication waveforms without CP, OFDM is a locally optimal waveform for QAM/PSK in the sense that it achieves a local minimum of the EISL of the A-ACF. Finally, we demonstrated by numerical results that under QAM/PSK constellations, there is no other orthogonal communication-centric waveform that achieves a lower ranging sidelobe level than that of the OFDM, in terms of both P-ACF and A-ACF cases.

cs.IT↗

arXiv2025

Kolmogorov-Arnold Networks-based GRU and LSTM for Loan Default Early Prediction

This study addresses a critical challenge in time series anomaly detection: enhancing the predictive capability of loan default models more than three months in advance to enable early identification of default events, helping financial institutions implement preventive measures before risk events materialize. Existing methods have significant drawbacks, such as their lack of accuracy in early predictions and their dependence on training and testing within the same year and specific time frames. These issues limit their practical use, particularly with out-of-time data. To address these, the study introduces two innovative architectures, GRU-KAN and LSTM-KAN, which merge Kolmogorov-Arnold Networks (KAN) with Gated Recurrent Units (GRU) and Long Short-Term Memory (LSTM) networks. The proposed models were evaluated against the baseline models (LSTM, GRU, LSTM-Attention, and LSTM-Transformer) in terms of accuracy, precision, recall, F1 and AUC in different lengths of feature window, sample sizes, and early prediction intervals. The results demonstrate that the proposed model achieves a prediction accuracy of over 92% three months in advance and over 88% eight months in advance, significantly outperforming existing baselines.

cs.LG↗

arXiv2025

Quantize More, Lose Less: Autoregressive Generation from Residually Quantized Speech Representations

Text-to-speech (TTS) synthesis has seen renewed progress under the discrete modeling paradigm. Existing autoregressive approaches often rely on single-codebook representations, which suffer from significant information loss. Even with post-hoc refinement techniques such as flow matching, these methods fail to recover fine-grained details (e.g., prosodic nuances, speaker-specific timbres), especially in challenging scenarios like singing voice or music synthesis. We propose QTTS, a novel TTS framework built upon our new audio codec, QDAC. The core innovation of QDAC lies in its end-to-end training of an ASR-based auto-regressive network with a GAN, which achieves superior semantic feature disentanglement for scalable, near-lossless compression. QTTS models these discrete codes using two innovative strategies: the Hierarchical Parallel architecture, which uses a dual-AR structure to model inter-codebook dependencies for higher-quality synthesis, and the Delay Multihead approach, which employs parallelized prediction with a fixed delay to accelerate inference speed. Our experiments demonstrate that the proposed framework achieves higher synthesis quality and better preserves expressive content compared to baseline. This suggests that scaling up compression via multi-codebook modeling is a promising direction for high-fidelity, general-purpose speech and audio generation.

cs.SD↗

arXiv2025

ED$^4$: Explicit Data-level Debiasing for Deepfake Detection

Learning intrinsic bias from limited data has been considered the main reason for the failure of deepfake detection with generalizability. Apart from the discovered content and specific-forgery bias, we reveal a novel spatial bias, where detectors inertly anticipate observing structural forgery clues appearing at the image center, also can lead to the poor generalization of existing methods. We present ED$^4$, a simple and effective strategy, to address aforementioned biases explicitly at the data level in a unified framework rather than implicit disentanglement via network design. In particular, we develop ClockMix to produce facial structure preserved mixtures with arbitrary samples, which allows the detector to learn from an exponentially extended data distribution with much more diverse identities, backgrounds, local manipulation traces, and the co-occurrence of multiple forgery artifacts. We further propose the Adversarial Spatial Consistency Module (AdvSCM) to prevent extracting features with spatial bias, which adversarially generates spatial-inconsistent images and constrains their extracted feature to be consistent. As a model-agnostic debiasing strategy, ED$^4$ is plug-and-play: it can be integrated with various deepfake detectors to obtain significant benefits. We conduct extensive experiments to demonstrate its effectiveness and superiority over existing deepfake detection approaches.

cs.CV↗

arXiv2025

Efficient Temporal Simple Path Graph Generation

Interactions between two entities often occur at specific timestamps, which can be modeled as a temporal graph. Exploring the relationships between vertices based on temporal paths is one of the fundamental tasks. In this paper, we conduct the first research to propose and investigate the problem of generating the temporal simple path graph (tspG), which is the subgraph consisting of all temporal simple paths from the source vertex to the target vertex within the given time interval. Directly enumerating all temporal simple paths and constructing the tspG is computationally expensive. To accelerate the processing, we propose an efficient method named Verification in Upper-bound Graph. It first incorporates the temporal path constraint and simple path constraint to exclude unpromising edges from the original graph, which obtains a tight upper-bound graph as a high-quality approximation of the tspG in polynomial time. Then, an Escape Edges Verification algorithm is further applied in the upper-bound graph to construct the exact tspG without exhaustively enumerating all temporal simple paths between given vertices. Finally, comprehensive experiments on 10 real-world graphs are conducted to demonstrate the efficiency and effectiveness of the proposed techniques.

cs.DB↗

arXiv2025

Quantum Latin squares with all possible cardinalities

A quantum Latin square of order $n$ (denoted as QLS$(n)$) is an $n\times n$ array whose entries are unit column vectors from the $n$-dimensional Hilbert space $\mathcal{H}_n$, such that each row and column forms an orthonormal basis. Two unit vectors $|u\rangle, |v\rangle\in \mathcal{H}_n$ are regarded as identical if there exists a real number $θ$ such that $|u\rangle=e^{iθ}|v\rangle$; otherwise, they are considered distinct. The cardinality $c$ of a QLS$(n)$ is the number of distinct vectors in the array. In this paper, we use sub-QLS$(4)$s to prove that for any integer $m\geq 2$ and any integer $c\in [4m,16m^2]\setminus \{4m+1\}$, there is a QLS$(4m)$ with cardinality $c$.

quant-ph↗

arXiv2025

A Factorized Mass Structure of Fermions and Its Fit

The structure of the mass matrix, a challenging problem in the Standard Model, is closely related to flavor phenomenology and the understanding of the Yukawa interaction. We derive a factorized mass structure based on observed fermion mass hierarchies, investigating the role of $SO(2)^f$ family symmetry in explaining the approximate degeneracy of light quark generations and its connection to flavor mixing. Our calculation includes the slightest modification from the lightest fermions by $\mathcal{O}(h^2)$ hierarchy corrections. Using this model-independent framework, we systematically analyze all mass patterns and perform comprehensive fits to both quark CKM and lepton PMNS mixing data with extended normal-ordered Dirac neutrinos. The results demonstrate this framework's capacity to unify flavor phenomena while simultaneously providing new insights into the fundamental nature of Yukawa interactions.

hep-ph↗

arXiv2025

$Z_c(3900)$ in a hadronic molecule and a triangle singularity approach at finite temperature

Studying exotic hadrons is a challenge against the conventional quark model, providing us with a good platform to deepen our understanding of the strong interaction. An inclusive study of the exotic hadrons in vacuum and at finite temperature is an intriguing approach to shed light on their nature. As a first step, we study the $Z_c(3900)$ in both the $D\bar{D}^\ast$ hadronic molecular and the triangle singularity pictures, and discuss the behaviors of $Z_c(3900)$ at finite temperature in these two different pictures. As a result, we show the properties that its mass becomes smaller and its width becomes larger when the temperature increases, which are seen commonly in the hadronic molecular picture and in the triangle singularity picture. The enhanced widths in hot medium indicate that the $Z_c(3900)$ will be dissociated at a sufficiently high temperature. This feature is also reflected by a decrease of the effective couplings in the hadronic molecular picture. It is concluded that, not only in vacuum but also in hot medium, the behaviors of $Z_{c}(3900)$ are similar in the two different interpretations of $Z_c(3900)$.

hep-ph↗

arXiv2025

Code Graph Model (CGM): A Graph-Integrated Large Language Model for Repository-Level Software Engineering Tasks

Recent advances in Large Language Models (LLMs) have shown promise in function-level code generation, yet repository-level software engineering tasks remain challenging. Current solutions predominantly rely on proprietary LLM agents, which introduce unpredictability and limit accessibility, raising concerns about data privacy and model customization. This paper investigates whether open-source LLMs can effectively address repository-level tasks without requiring agent-based approaches. We demonstrate this is possible by enabling LLMs to comprehend functions and files within codebases through their semantic information and structural dependencies. To this end, we introduce Code Graph Models (CGMs), which integrate repository code graph structures into the LLM's attention mechanism and map node attributes to the LLM's input space using a specialized adapter. When combined with an agentless graph RAG framework, our approach achieves a 43.00% resolution rate on the SWE-bench Lite benchmark using the open-source Qwen2.5-72B model. This performance ranks first among open weight models, second among methods with open-source systems, and eighth overall, surpassing the previous best open-source model-based method by 12.33%.

cs.SE↗

arXiv2025

WIP: Large Language Model-Enhanced Smart Tutor for Undergraduate Circuit Analysis

This research-to-practice work-in-progress (WIP) paper presents an AI-enabled smart tutor designed to provide homework assessment and feedback for students in an undergraduate circuit analysis course. We detail the tutor's design philosophy and core components, including open-ended question answering and homework feedback generation. The prompts are carefully crafted to optimize responses across different problems. The smart tutor was deployed on the Microsoft Azure platform and is currently in use in an undergraduate circuit analysis course at the School of Electrical and Computer Engineering in a large, public, research-intensive institution in the Southeastern United States. Beyond offering personalized instruction and feedback, the tutor collects student interaction data, which is summarized and shared with the course instructor. To evaluate its effectiveness, we collected student feedback, with 90.9% of responses indicating satisfaction with the tutor. Additionally, we analyze a subset of collected data on preliminary circuit analysis topics to assess tutor usage frequency for each problem and identify frequently asked questions. These insights help instructors gain real-time awareness of student difficulties, enabling more targeted classroom instruction. In future work, we will release a full analysis once the complete dataset is available after the Spring 2025 semester. We also explore the potential applications of this smart tutor across a broader range of engineering disciplines by developing improved prompts, diagram-recognition methods, and database management strategies, which remain ongoing areas of research.

cs.CY↗

arXiv2025

Diversity of Transformer Layers: One Aspect of Parameter Scaling Laws

Transformers deliver outstanding performance across a wide range of tasks and are now a dominant backbone architecture for large language models (LLMs). Their task-solving performance is improved by increasing parameter size, as shown in the recent studies on parameter scaling laws. Although recent mechanistic-interpretability studies have deepened our understanding of the internal behavior of Transformers by analyzing their residual stream, the relationship between these internal mechanisms and the parameter scaling laws remains unclear. To bridge this gap, we focus on layers and their size, which mainly decide the parameter size of Transformers. For this purpose, we first theoretically investigate the layers within the residual stream through a bias-diversity decomposition. The decomposition separates (i) bias, the error of each layer's output from the ground truth, and (ii) diversity, which indicates how much the outputs of each layer differ from each other. Analyzing Transformers under this theory reveals that performance improves when individual layers make predictions close to the correct answer and remain mutually diverse. We show that diversity becomes especially critical when individual layers' outputs are far from the ground truth. Finally, we introduce an information-theoretic diversity and show our main findings that adding layers enhances performance only when those layers behave differently, i.e., are diverse. We also reveal the performance gains from increasing the number of layers exhibit submodularity: marginal improvements diminish as additional layers increase, mirroring the logarithmic convergence predicted by the parameter scaling laws. Experiments on multiple semantic-understanding tasks with various LLMs empirically confirm the theoretical properties derived in this study.

cs.CL↗