YZYexun Zhang

Yexun Zhang

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Yexun Zhang has a total of 186 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Dual-granularity Sinkhorn Distillation for Enhanced Learning from Long-tailed Noisy Data

    by Feng Hong, Yu Huang, Zihua Zhao, Zhihan Zhou, Jiangchao Yao, Dongsheng Li, Ya Zhang, Yanfeng Wang

    Real-world datasets for deep learning frequently suffer from the co-occurring challenges of class imbalance and label noise, hindering model performance. While methods exist for each issue, effectively combining them is non-trivial, as distinguishing genuine tail samples from noisy data proves difficult, often leading to conflicting optimization strategies. This paper presents a novel perspective: instead of primarily developing new complex techniques from scratch, we explore synergistically leveraging well-established, individually 'weak' auxiliary models - specialized for tackling either class imbalance or label noise but not both. This view is motivated by the insight that class imbalance (a distributional-level concern) and label noise (a sample-level concern) operate at different granularities, suggesting that robustness mechanisms for each can in principle offer complementary strengths without conflict. We propose Dual-granularity Sinkhorn Distillation (D-SINK), a novel framework that enhances dual robustness by distilling and integrating complementary insights from such 'weak', single-purpose auxiliary models. Specifically, D-SINK uses an optimal transport-optimized surrogate label allocation to align the target model's sample-level predictions with a noise-robust auxiliary and its class distributions with an imbalance-robust one. Extensive experiments on benchmark datasets demonstrate that D-SINK significantly improves robustness and achieves strong empirical performance in learning from long-tailed noisy data.

  • Open Access

    ResAD++: Towards Class Agnostic Anomaly Detection via Residual Feature Learning

    by Xincheng Yao, Chao Shi, Muming Zhao, Guangtao Zhai, Chongyang Zhang

    This paper explores the problem of class-agnostic anomaly detection (AD), where the objective is to train one class-agnostic AD model that can generalize to detect anomalies in diverse new classes from different domains without any retraining or fine-tuning on the target data. When applied for new classes, the performance of current single- and multi-class AD methods is still unsatisfactory. One fundamental reason is that representation learning in existing methods is still class-related, namely, feature correlation. To address this issue, we propose residual features and construct a simple but effective framework, termed ResAD. Our core insight is to learn the residual feature distribution rather than the initial feature distribution. Residual features are formed by matching and then subtracting normal reference features. In this way, we can effectively realize feature decorrelation. Even in new classes, the distribution of normal residual features would not remarkably shift from the learned distribution. In addition, we think that residual features still have one issue: scale correlation. To this end, we propose a feature hypersphere constraining approach, which learns to constrain initial normal residual features into a spatial hypersphere for enabling the feature scales of different classes as consistent as possible. Furthermore, we propose a novel logbarrier bidirectional contraction OCC loss and vector quantization based feature distribution matching module to enhance ResAD, leading to the improved version of ResAD (ResAD++). Comprehensive experiments on eight real-world AD datasets demonstrate that our ResAD++ can achieve remarkable AD results when directly used in new classes, outperforming state-of-the-art competing methods and also surpassing ResAD. The code is available at https://github.com/xcyao00/ResAD.

  • Open Access

    The Tenth NTIRE 2025 Efficient Super-Resolution Challenge Report

    by Bin Ren, Hang Guo, Lei Sun, Zongwei Wu, Radu Timofte, Yawei Li, Yao Zhang, Xinning Chai, Zhengxue Cheng, Yingsheng Qin, et al.

    This paper presents a comprehensive review of the NTIRE 2025 Challenge on Single-Image Efficient Super-Resolution (ESR). The challenge aimed to advance the development of deep models that optimize key computational metrics, i.e., runtime, parameters, and FLOPs, while achieving a PSNR of at least 26.90 dB on the $\operatorname{DIV2K\_LSDIR\_valid}$ dataset and 26.99 dB on the $\operatorname{DIV2K\_LSDIR\_test}$ dataset. A robust participation saw \textbf{244} registered entrants, with \textbf{43} teams submitting valid entries. This report meticulously analyzes these methods and results, emphasizing groundbreaking advancements in state-of-the-art single-image ESR techniques. The analysis highlights innovative approaches and establishes benchmarks for future research in the field.

  • Open Access

    Towards Diverse Temporal Grounding under Single Positive Labels

    by Hao Zhou, Chongyang Zhang, Yanjun Chen, Chuanping Hu

    Temporal grounding aims to retrieve moments of the described event within an untrimmed video by a language query. Typically, existing methods assume annotations are precise and unique, yet one query may describe multiple moments in many cases. Hence, simply taking it as a one-vs-one mapping task and striving to match single-label annotations will inevitably introduce false negatives during optimization. In this study, we reformulate this task as a one-vs-many optimization problem under the condition of single positive labels. The unlabeled moments are considered unobserved rather than negative, and we explore mining potential positive moments to assist in multiple moment retrieval. In this setting, we propose a novel Diverse Temporal Grounding framework, termed DTG-SPL, which mainly consists of a positive moment estimation (PME) module and a diverse moment regression (DMR) module. PME leverages semantic reconstruction information and an expected positive regularization to uncover potential positive moments in an online fashion. Under the supervision of these pseudo positives, DMR is able to localize diverse moments in parallel that meet different users. The entire framework allows for end-to-end optimization as well as fast inference. Extensive experiments on Charades-STA and ActivityNet Captions show that our method achieves superior performance in terms of both single-label and multi-label metrics.

  • Open Access

    FedSkip: Combatting Statistical Heterogeneity with Federated Skip Aggregation

    by Ziqing Fan, Yanfeng Wang, Jiangchao Yao, Lingjuan Lyu, Ya Zhang, Qi Tian

    The statistical heterogeneity of the non-independent and identically distributed (non-IID) data in local clients significantly limits the performance of federated learning. Previous attempts like FedProx, SCAFFOLD, MOON, FedNova and FedDyn resort to an optimization perspective, which requires an auxiliary term or re-weights local updates to calibrate the learning bias or the objective inconsistency. However, in addition to previous explorations for improvement in federated averaging, our analysis shows that another critical bottleneck is the poorer optima of client models in more heterogeneous conditions. We thus introduce a data-driven approach called FedSkip to improve the client optima by periodically skipping federated averaging and scattering local models to the cross devices. We provide theoretical analysis of the possible benefit from FedSkip and conduct extensive experiments on a range of datasets to demonstrate that FedSkip achieves much higher accuracy, better aggregation efficiency and competing communication efficiency. Source code is available at: https://github.com/MediaBrain-SJTU/FedSkip.

  • Open Access

    Degeneration in VAE: in the Light of Fisher Information Loss

    by Huangjie Zheng, Jiangchao Yao, Ya Zhang, Ivor W. Tsang

    While enormous progress has been made to Variational Autoencoder (VAE) in recent years, similar to other deep networks, VAE with deep networks suffers from the problem of degeneration, which seriously weakens the correlation between the input and the corresponding latent codes, deviating from the goal of the representation learning. To investigate how degeneration affects VAE from a theoretical perspective, we illustrate the information transmission in VAE and analyze the intermediate layers of the encoders/decoders. Specifically, we propose a Fisher Information measure for the layer-wise analysis. With such measure, we demonstrate that information loss is ineluctable in feed-forward networks and causes the degeneration in VAE. We show that skip connections in VAE enable the preservation of information without changing the model architecture. We call this class of VAE equipped with skip connections as SCVAE and perform a range of experiments to show its advantages in information preservation and degeneration mitigation.

  • Open Access

    Separating Style and Content for Generalized Style Transfer

    by Yexun Zhang, Ya Zhang, Wenbin Cai, Jie Chang

    Neural style transfer has drawn broad attention in recent years. However, most existing methods aim to explicitly model the transformation between different styles, and the learned model is thus not generalizable to new styles. We here attempt to separate the representations for styles and contents, and propose a generalized style transfer network consisting of style encoder, content encoder, mixer and decoder. The style encoder and content encoder are used to extract the style and content factors from the style reference images and content reference images, respectively. The mixer employs a bilinear model to integrate the above two factors and finally feeds it into a decoder to generate images with target style and content. To separate the style features and content features, we leverage the conditional dependence of styles and contents given an image. During training, the encoder network learns to extract styles and contents from two sets of reference images in limited size, one with shared style and the other with shared content. This learning framework allows simultaneous style transfer among multiple styles and can be deemed as a special `multi-task' learning scenario. The encoders are expected to capture the underlying features for different styles and contents which is generalizable to new styles and contents. For validation, we applied the proposed algorithm to the Chinese Typeface transfer problem. Extensive experiment results on character generation have demonstrated the effectiveness and robustness of our method.


  • Open Access

    Enhancing the biocatalytic manufacture of the key intermediate of atorvastatin by focused directed evolution of halohydrin dehalogenase

    by Yu Luo, Yangzi Chen, Hongmin Ma, ZhenHua Tian, Yeqi Zhang, Jian Zhang

    Halohydrin dehalogenases (HHDHs) are biocatalytically interesting enzymes due to their ability to form C-C, C-N, C-O, and C-S bonds. One of most important application of HHDH was the protein engineering of HheC (halohydrin dehalogenase from Agrobacterium radiobacter AD1) for the industrial manufacturing of ethyl (R)-4-cyano-3-hydroxybutanoate (HN), a key chiral synthon of a cholesterol-lowering drug of atorvastatin. During our development of an alternative, more efficient and economic route for chemo-enzymatic preparation of the intermediate of atorvastatin, we found that the HheC2360 previously reported for HN manufacture, had insufficient activity for the cyanolysis production of tert-butyl (3 R,5 S)-6-cyano-3,5-dihydroxyhexanoate (A7). Herein, we present the focused directed evolution of HheC2360 with higher activity and enhanced biocatalytic performance using active site mutagenesis. Through docking of the product, A7, into the crystal structure of HheC2360, 6 residues was selected for combined active sites testing (CASTing). After library screening, the variant V84G/W86F was identified to have a 15- fold increase in activity. Time course analysis of the cyanolysis reaction catalyzed by this variant, showed 2- fold increase in space time productivity compared with HheC2360. These results demonstrate the applicability of the variant V84G/W86F as a biocatalyst for the efficient and practical production of atorvastatin intermediate.


  • Open Access

    Dynamic Clustering in Suspension of Motile Bacteria

    by Xiao Chen, Xiang Yang, Mingcheng Yang, Hepeng Zhang

    Bacteria suspension exhibits a wide range of collective phenomena arising from interactions between individual cells. Here we show Serratia marcescens cells near an air-liquid interface spontaneously aggregate into dynamic clusters through surface-mediated hydrodynamic interactions. These long-lived clusters translate randomly and rotate in the counter-clockwise direction; they continuously evolve, merge with others and split into smaller ones. Measurements indicate that long-ranged hydrodynamic interactions have strong influences on cluster properties. Bacterial clusters change material and fluid transport near the interface and hence may have environmental and biological consequences.

  • Open Access

    Back Cover: Two-Dimensional Sandwich-Type, Graphene-Based Conjugated Microporous Polymers (Angew. Chem. Int. Ed. 37/2013)

    by Xiaodong Zhuang, Fan Zhang, Dongqing Wu, Nina Forler, Haiwei Liang, Manfred Wagner, Dominik Gehrig, Michael Ryan Hansen, Frédéric Laquai, Xinliang Feng