Author name cluster

Yu Deng

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.

10 papers

1 author row

EAAI Journal 2026 Journal Article

A deep learning-based imaging classification framework for interstitial lung disease

Hongyi Wang
Anqi Liu
Xiaoyan Yang
Yifei Ni
Jianping Wang
Jie Du
Yuhui Qiang
Bingbing Xie

Rationale Interstitial lung diseases (ILD) are a diverse group of conditions, often diagnosed using high-resolution chest computed tomography (HRCT), which is susceptible to subjective biases in interpretation. Objectives This study aims to develop and validate SPAIDNet (Spatial Pattern Analysis for ILD Diagnosis using a residual neural Network), a deep learning (DL) model for the automated classification of ILD, to reduce subjective biases and improve diagnostic consistency. Methods The study included 2901 ILD patients who underwent 5213 HRCT scans across multiple centers between July 2017 and June 2023. SPAIDNet, built upon the pre-trained residual neural network with 18 layers, utilizes multi-instance learning in three centers in China. Measurements and main results The model demonstrated exceptional performance, achieving macro-average area under the receiver operating characteristic curve (AUC) of over 0. 999 in internal validation, 0. 905 in external cohort I, and 0. 870 in external cohort II for multiclass classification. SPAIDNet outperformed both a junior radiologist (AUC: 0. 737) and a senior radiologist (AUC: 0. 763). Furthermore, DL-assisted the two radiologists saw significant improvements in diagnostic accuracy, with AUCs rising to 0. 817 and 0. 787, respectively. Conclusions These results underscore SPAIDNet's potential to offer high accuracy, robustness, and generalizability in ILD diagnosis, providing a valuable tool to mitigate the subjectivity inherent in HRCT image interpretation.

Details DOI

YNIMG Journal 2026 Journal Article

Foundation models for brain imaging: A systematic review

Salah Ghamizi
Georgia Kanli
Yu Deng
Valérie Palissot
Magali Perquin
Olivier Keunen

Foundation models (FMs), large neural networks pretrained on extensive and diverse datasets, have revolutionized artificial intelligence and demonstrated significant promise in medical imaging by enabling robust performance with limited labeled data. Although numerous surveys have reviewed the application of FMs in healthcare, brain imaging remains underrepresented, despite its critical role in the diagnosis and treatment of neurological diseases using modalities such as magnetic resonance imaging (MRI), computed tomography (CT), and positron emission tomography (PET). To address this gap, we present the first comprehensive and curated review of FMs for brain imaging. We systematically analyze 161 brain imaging datasets and 143 FMs up to Jan, 2026, providing insights into key design choices, training paradigms, and optimizations driving recent advances. Our review highlights that the race for larger models has stabilized in 2026 towards more efficient models. FMs for brain imaging heavily rely on MRI (92%) and CT (57%) inputs, while PET imaging remains vastly underexplored (supported by only 15% of models). Our study also demonstrates architectural vulnerabilities caused by homogenization and lack of diversity, with Vision Transformers utilized in 48% of visual encoders, and models predominantly built by patching pre-existing natural image backbones like SAM (19%), and CLIP (12%) rather than utilizing native domain-specific 3D medical imaging innovations. For each of the eight tasks of the study the systematic review identifies the best models and discusses their innovations. Our study also uncovers critical gaps in the tasks, pathologies and clinical validation. We demonstrate that the literature is disproportionately skewed toward brain cancer research (37% of models) and neurodegenerative diseases (24%), and discuss the potential causes and remedies. Similarly, tasks are heavily weighted toward anomaly classification (44%) and segmentation (32%), leaving areas like mental health and image synthesis underrepresented. Besides, most models rely exclusively on traditional machine learning metrics (e.g., DICE or SSIM) rather than medically relevant measures, and only seven out of the 143 models incorporated human expert evaluations to verify real-world utility. Our systematic review concludes by outlining future research directions to advance FMs in brain imaging and actionable recommendations to build better FMs and to evaluate and deploy them in clinical and research settings.

Details DOI

EAAI Journal 2025 Journal Article

A composite insulator aging level classification method based on fourier transform infrared spectroscopy and deep learning model

Zhenan Zhou
Chuyan Zhang
Kun Zhang
Yu Deng
Chen Gu
Yinan Lin
Xinzhe Yu

Composite insulators are critical components in power systems that ensure operational reliability, and accurate assessment of their aging status is crucial for grid safety. In this study, a large-scale dataset of Fourier Transform Infrared Spectroscopy (FTIR) data from methyl vinyl silicone rubber (MVQ) composite insulators is collected, and a classification model for aging levels is developed. An end-to-end composite insulator aging classification model based on deep learning, named Multiscale Depthwise Separable Convolution-Attention (MDSC-Attention), is proposed. The model effectively reduces computational complexity and improves efficiency by introducing one-dimensional depthwise separable convolutions. Additionally, it significantly enhances feature extraction and global modeling capabilities for FTIR data by combining multi-scale feature extraction modules with a self-attention mechanism. Experimental results demonstrate that the MDSC-Attention model can automatically extract key features from FTIR data and achieve high-precision classification of composite insulator aging levels, with a classification accuracy of 99. 68 % on the test set. This study validates the potential application of deep learning in composite material aging assessment and provides significant support for state monitoring and intelligent evaluation in power systems.

Details DOI

EAAI Journal 2025 Journal Article

An angle-enhanced deep learning framework for thermal defect diagnosis in overhead transmission line composite insulators

Xinzhe Yu
Zhenan Zhou
Yu Deng
Kun Zhang
Chen Gu
Zheyuan Liu
Songsong Zhou

Due to long-term outdoor exposure, composite insulators are susceptible to degradation and abnormal temperature rise, making Unmanned Aerial Vehicle (UAV)-based infrared inspections essential for effective monitoring. However, traditional manual interpretation of these images is inefficient and subjective. To improve detection automation and accuracy, we propose an intelligent detection method for composite insulators in infrared images based on an improved You Only Look Once version 11 (YOLOv11) model. The proposed approach introduces Oriented Bounding Boxes (OBBs) for annotation and designs an Angle-Enhanced Probabilistic Intersection over Union (AE-ProbIoU) loss function to enhance the model's ability to detect rotated objects. Experimental results demonstrate that the proposed Angle-Enhanced You Only Look Once (AE-YOLO) model achieves a mAP50: 95 of 94. 0 % and an angle prediction accuracy of 94. 3 %. In addition, a temperature extraction module based on the OBBs is developed to accurately derive the temperature profile of the insulator core rod. This method significantly enhances the intelligence level of infrared image analysis for composite insulators and provides technical support for condition assessment and fault prediction in power transmission lines.

Details DOI

NeurIPS Conference 2025 Conference Paper

MoGe-2: Accurate Monocular Geometry with Metric Scale and Sharp Details

Ruicheng Wang
Sicheng Xu
Yue Dong
Yu Deng
Jianfeng Xiang
Zelong Lv
Guangzhong Sun
Xin Tong

We propose MoGe-2, an advanced open-domain geometry estimation model that recovers a metric-scale 3D point map of a scene from a single image. Our method builds upon the recent monocular geometry estimation approach, MoGe, which predicts affine-invariant point maps with unknown scales. We explore effective strategies to extend MoGe for metric geometry prediction without compromising the relative geometry accuracy provided by the affine-invariant point representation. Additionally, we discover that noise and errors in real data diminish fine-grained detail in the predicted geometry. We address this by developing a data refinement approach that filters and completes real data using sharp synthetic labels, significantly enhancing the granularity of the reconstructed geometry while maintaining the overall accuracy. We train our model on a large corpus of mixed datasets and conducted comprehensive evaluations, demonstrating its superior performance in achieving accurate relative geometry, precise metric scale, and fine-grained detail recovery -- capabilities that no previous methods have simultaneously achieved.

PDF Details

NeurIPS Conference 2025 Conference Paper

STRATUS: A Multi-agent System for Autonomous Reliability Engineering of Modern Clouds

Yinfang Chen
Jiaqi Pan
Jackson Clark
Yiming Su
Noah Zheutlin
Bhavya Bhavya
Rohan R. Arora
Yu Deng

In cloud-scale systems, failures are the norm. A distributed computing cluster exhibits hundreds of machine failures and thousands of disk failures; software bugs and misconfigurations are reported to be more frequent. The demand for autonomous, AI-driven reliability engineering continues to grow, as existing human-in-the-loop practices can hardly keep up with the scale of modern clouds. This paper presents STRATUS, an LLM-based multi-agent system for realizing autonomous Site Reliability Engineering (SRE) of cloud services. STRATUS consists of multiple specialized agents (e. g. , for failure detection, diagnosis, mitigation), organized in a state machine to assist system-level safety reasoning and enforcement. We formalize a key safety specification of agentic SRE systems like STRATUS, termed Transactional No-Regression (TNR), which enables safe exploration and iteration. We show that TNR can effectively improve autonomous failure mitigation. STRATUS significantly outperforms state-of-the-art SRE agents in terms of success rate of failure mitigation problems in AIOpsLab and ITBench (two SRE benchmark suites), by at least 1. 5 times across various models. STRATUS shows a promising path toward practical deployment of agentic systems for cloud reliability.

PDF Details

NeurIPS Conference 2025 Conference Paper

VASA-3D: Lifelike Audio-Driven Gaussian Head Avatars from a Single Image

Sicheng Xu
Guojun Chen
Jiaolong Yang
Yizhong Zhang
Yu Deng
Stephen Lin
Baining Guo

We propose VASA-3D, an audio-driven, single-shot 3D head avatar generator. This research tackles two major challenges: capturing the subtle expression details present in real human faces, and reconstructing an intricate 3D head avatar from a single portrait image. To accurately model expression details, VASA-3D leverages the motion latent of VASA-1, a method that yields exceptional realism and vividness in 2D talking heads. A critical element of our work is translating this motion latent to 3D, which is accomplished by devising a 3D head model that is conditioned on the motion latent. Customization of this model to a single image is achieved through an optimization framework that employs numerous video frames of the reference head synthesized from the input image. The optimization takes various training losses robust to artifacts and limited pose coverage in the generated training data. Our experiment shows that VASA-3D produces realistic 3D talking heads that cannot be achieved by prior art, and it supports the online generation of 512x512 free-viewpoint videos at up to 75 FPS, facilitating more immersive engagements with lifelike 3D avatars.

PDF Details

AAAI Conference 2024 Conference Paper

Seed-Guided Fine-Grained Entity Typing in Science and Engineering Domains

Yu Zhang
Yunyi Zhang
Yanzhen Shen
Yu Deng
Lucian Popa
Larisa Shwartz
ChengXiang Zhai
Jiawei Han

Accurately typing entity mentions from text segments is a fundamental task for various natural language processing applications. Many previous approaches rely on massive human-annotated data to perform entity typing. Nevertheless, collecting such data in highly specialized science and engineering domains (e.g., software engineering and security) can be time-consuming and costly, without mentioning the domain gaps between training and inference data if the model needs to be applied to confidential datasets. In this paper, we study the task of seed-guided fine-grained entity typing in science and engineering domains, which takes the name and a few seed entities for each entity type as the only supervision and aims to classify new entity mentions into both seen and unseen types (i.e., those without seed entities). To solve this problem, we propose SEType which first enriches the weak supervision by finding more entities for each seen type from an unlabeled corpus using the contextualized representations of pre-trained language models. It then matches the enriched entities to unlabeled text to get pseudo-labeled samples and trains a textual entailment model that can make inferences for both seen and unseen types. Extensive experiments on two datasets covering four domains demonstrate the effectiveness of SEType in comparison with various baselines. Code and data are available at: https://github.com/yuzhimanhua/SEType.

PDF Details DOI

NeurIPS Conference 2022 Conference Paper

AniFaceGAN: Animatable 3D-Aware Face Image Generation for Video Avatars

Yue Wu
Yu Deng
Jiaolong Yang
Fangyun Wei
Qifeng Chen
Xin Tong

Although 2D generative models have made great progress in face image generation and animation, they often suffer from undesirable artifacts such as 3D inconsistency when rendering images from different camera viewpoints. This prevents them from synthesizing video animations indistinguishable from real ones. Recently, 3D-aware GANs extend 2D GANs for explicit disentanglement of camera pose by leveraging 3D scene representations. These methods can well preserve the 3D consistency of the generated images across different views, yet they cannot achieve fine-grained control over other attributes, among which facial expression control is arguably the most useful and desirable for face animation. In this paper, we propose an animatable 3D-aware GAN for multiview consistent face animation generation. The key idea is to decompose the 3D representation of the 3D-aware GAN into a template field and a deformation field, where the former represents different identities with a canonical expression, and the latter characterizes expression variations of each identity. To achieve meaningful control over facial expressions via deformation, we propose a 3D-level imitative learning scheme between the generator and a parametric 3D face model during adversarial training of the 3D-aware GAN. This helps our method achieve high-quality animatable face image generation with strong visual 3D consistency, even though trained with only unstructured 2D images. Extensive experiments demonstrate our superior performance over prior works. Project page: \url{https: //yuewuhkust. github. io/AniFaceGAN/

PDF Details

AAAI Conference 2019 Conference Paper

Validation of Growing Knowledge Graphs by Abductive Text Evidences

Jianfeng Du
Jeff Z. Pan
Sylvia Wang
Kunxun Qi
Yuming Shen
Yu Deng

This paper proposes a validation mechanism for newly added triples in a growing knowledge graph. Given a logical theory, a knowledge graph, a text corpus, and a new triple to be validated, this mechanism computes a sorted list of explanations for the new triple to facilitate the validation of it, where an explanation, called an abductive text evidence, is a set of pairs of the form (triple, window) where appending the set of triples on the left to the knowledge graph enforces entailment of the new triple under the logical theory, while every sentence window on the right which is contained in the text corpus explains to some degree why the triple on the left is true. From the angle of practice, a special class of abductive text evidences called TEP-based abductive text evidence is proposed, which is constructed from explanation patterns seen before in the knowledge graph. Accordingly, a method for computing the complete set of TEP-based abductive text evidences is proposed. Moreover, a method for sorting abductive text evidences based on distantly supervised learning is proposed. To evaluate the proposed validation mechanism, four knowledge graphs with logical theories are constructed from the four great classical masterpieces of Chinese literature. Experimental results on these datasets demonstrate the efficiency and effectiveness of the proposed mechanism.

PDF Details