no code implementations • 23 May 2024 • Kai Yao, Zhaorui Tan, Zixian Su, Xi Yang, Jie Sun, Kaizhu Huang
Built upon this, we argue that conventional OCDA approaches may substantially underestimate the inherent variance inside the compound target domains for model generalization.
no code implementations • 14 Apr 2024 • Yuqi Wang, Zeqiang Wang, Wei Wang, Qi Chen, Kaizhu Huang, Anh Nguyen, Suparna De
Safe and reliable natural language inference is critical for extracting insights from clinical trial reports but poses challenges due to biases in large pre-trained language models.
no code implementations • 28 Mar 2024 • Tianyi Liu, Zhaorui Tan, Kaizhu Huang, Haochuan Jiang
Medical image segmentation presents the challenge of segmenting various-size targets, demanding the model to effectively capture both local and global information.
1 code implementation • 29 Feb 2024 • Zhaorui Tan, Xi Yang, Kaizhu Huang
Multi-domain generalization (mDG) is universally aimed to minimize the discrepancy between training and testing distributions to enhance marginal-to-label distribution mapping.
no code implementations • 17 Jan 2024 • Jingwei Guo, Kaizhu Huang, Xinping Yi, Zixian Su, Rui Zhang
Whilst spectral Graph Neural Networks (GNNs) are theoretically well-founded in the spectral domain, their practical reliance on polynomial approximation implies a profound linkage to the spatial domain.
no code implementations • 21 Dec 2023 • Jing Li, Qiu-Feng Wang, Kaizhu Huang, Rui Zhang, Siyuan Wang
To train Diff-Oracle effectively, we propose to obtain pixel-level paired oracle character images (i. e., style and content images) by a pre-trained image-to-image translation model.
1 code implementation • 15 Dec 2023 • Zixian Su, Jingwei Guo, Kai Yao, Xi Yang, Qiufeng Wang, Kaizhu Huang
While recent test-time adaptations exhibit efficacy by adjusting batch normalization to narrow domain disparities, their effectiveness diminishes with realistic mini-batches due to inaccurate target estimation.
1 code implementation • 14 Dec 2023 • Jingwei Guo, Kaizhu Huang, Xinping Yi, Rui Zhang
Spectral Graph Neural Networks (GNNs) have achieved tremendous success in graph machine learning, with polynomial filters applied for graph convolutions, where all nodes share the identical filter weights to mine their local contexts.
no code implementations • 13 Dec 2023 • Weiguang Zhang, Qiufeng Wang, Kaizhu Huang
While Cartesian coordinates are typically leveraged by state-of-the-art approaches to learn a group of deformation control points, such representation is not efficient for dewarping model to learn the deformation information.
1 code implementation • 13 Dec 2023 • Zhaorui Tan, Xi Yang, Kaizhu Huang
In particular, we propose to augment texts in the semantic space via an Implicit Textual Semantic Preserving Augmentation ($ITA$), in conjunction with a specifically designed Image Semantic Regularization Loss ($L_r$) as Generated Image Semantic Conservation, to cope well with semantic mismatch and collapse.
2 code implementations • 12 Dec 2023 • Weiguang Zhao, Guanyu Yang, Rui Zhang, Chenru Jiang, Chaolong Yang, Yuyao Yan, Amir Hussain, Kaizhu Huang
To this end, we propose a more realistic and challenging scenario named open-pose 3D zero-shot classification, focusing on the recognition of 3D objects regardless of their orientation.
no code implementations • 31 Oct 2023 • Yuqi Wang, Zeqiang Wang, Wei Wang, Qi Chen, Kaizhu Huang, Anh Nguyen, Suparna De
In the era of the Internet of Things (IoT), the retrieval of relevant medical information has become essential for efficient clinical decision-making.
no code implementations • 25 Oct 2023 • Yiming Lin, Xiao-Bo Jin, Qiufeng Wang, Kaizhu Huang
The current state-of-the-art methods first refine the representation of phrase by aggregating the most similar $k$ image pixels, and then match the refined text representations with the pixels of the image feature map to generate segmentation results.
no code implementations • 4 Sep 2023 • ZiHao Zhou, Qiufeng Wang, Mingyu Jin, Jie Yao, Jianan Ye, Wei Liu, Wei Wang, Xiaowei Huang, Kaizhu Huang
Instead of attacking prompts in the use of LLMs, we propose a MathAttack model to attack MWP samples which are closer to the essence of security in solving math problems.
1 code implementation • 5 Aug 2023 • Maizhen Ning, Qiu-Feng Wang, Kaizhu Huang, Xiaowei Huang
For the diagram encoder, we pre-train it under a multi-label classification framework with the symbolic characters as labels.
1 code implementation • 15 Jun 2023 • ZiHao Zhou, Maizhen Ning, Qiufeng Wang, Jie Yao, Wei Wang, Xiaowei Huang, Kaizhu Huang
We then feed them to a question generator together with the scenario to obtain the corresponding diverse questions, forming a new MWP with a variety of questions and equations.
no code implementations • 14 Jun 2023 • Jianan Ye, Yijie Hu, Xi Yang, Qiu-Feng Wang, Chao Huang, Kaizhu Huang
We then design a novel patch-wise residual module in the anomaly learning head to extract and assess the fine-grained anomaly features from each sample, facilitating the learning of discriminative representations of anomaly instances.
no code implementations • 20 Apr 2023 • Jiezhu Cheng, Kaizhu Huang, Zibin Zheng
By lowering the volatility of the stock recommendation model, SVAT effectively reduces investment risks and outperforms state-of-the-art baselines by more than 30% in terms of risk-adjusted profits.
no code implementations • 12 Feb 2023 • Shiran Yuan, Kaizhu Huang
This work attempts to construct a new methodological framework called GCDTC (Generalized CP Decomposition Tensor Completion) for leveraging numerical priors and achieving higher accuracy in tensor completion.
1 code implementation • ICCV 2023 • Zhiqiang Gao, Kaizhu Huang, Rui Zhang, Dawei Liu, Jieming Ma
Recent studies have investigated how to achieve robustness for unsupervised domain adaptation (UDA).
no code implementations • 13 Dec 2022 • Chaolong Yang, Yuyao Yan, Weiguang Zhao, Jianan Ye, Xi Yang, Amir Hussain, Kaizhu Huang
On the one hand, the unidirectional projection enforces our model focused more on the core task, i. e., 3D segmentation; on the other hand, unlocking the bidirectional to unidirectional projection enables a deeper cross-domain semantic alignment and enjoys the flexibility to fuse better and complicated features from very different spaces.
no code implementations • 7 Dec 2022 • M. Tanveer, M. A. Ganaie, Iman Beheshti, Tripti Goel, Nehal Ahmad, Kuan-Ting Lai, Kaizhu Huang, Yu-Dong Zhang, Javier Del Ser, Chin-Teng Lin
In this review, we offer a comprehensive analysis of the literature related to the adoption of deep learning for brain age estimation with neuroimaging data.
1 code implementation • 27 Nov 2022 • Zixian Su, Kai Yao, Xi Yang, Qiufeng Wang, Jie Sun, Kaizhu Huang
Single-source domain generalization (SDG) in medical image segmentation is a challenging yet essential task as domain shifts are quite common among clinical image datasets.
1 code implementation • 27 Oct 2022 • Zhaorui Tan, Xi Yang, Zihan Ye, Qiufeng Wang, Yuyao Yan, Anh Nguyen, Kaizhu Huang
Generating consistent and high-quality images from given texts is essential for visual-language understanding.
1 code implementation • 13 Oct 2022 • Zihan Ye, Guanyu Yang, Xiaobo Jin, Youfa Liu, Kaizhu Huang
Broadly speaking, present ZSL methods usually adopt class-level semantic labels and compare them with instance-level semantic predictions to infer unseen classes.
1 code implementation • 3 Aug 2022 • Penglei Gao, Xi Yang, Rui Zhang, Ping Guo, John Y. Goulermas, Kaizhu Huang
While exogenous variables have a major impact on performance improvement in time series analysis, inter-series correlation and time dependence among them are rarely considered in the present continuous methods.
1 code implementation • ICCV 2023 • Weiguang Zhao, Yuyao Yan, Chaolong Yang, Jianan Ye, Xi Yang, Kaizhu Huang
Due to the uneven distribution of offset points, these existing methods can hardly cluster all instance points.
Ranked #3 on 3D Instance Segmentation on S3DIS
1 code implementation • 12 Jul 2022 • Kai Yao, Penglei Gao, Xi Yang, Kaizhu Huang, Jie Sun, Rui Zhang
Image outpainting, which is well studied with Convolution Neural Network (CNN) based framework, has recently drawn more attention in computer vision.
1 code implementation • 27 May 2022 • Jingwei Guo, Kaizhu Huang, Rui Zhang, Xinping Yi
While Graph Neural Networks (GNNs) have achieved enormous success in multiple graph analytical tasks, modern variants mostly rely on the strong inductive bias of homophily.
no code implementations • 24 May 2022 • Zixian Su, Kai Yao, Xi Yang, Qiufeng Wang, Yuyao Yan, Jie Sun, Kaizhu Huang
This combination of global and local alignment can precisely localize the crucial regions in segmentation target while preserving the overall semantic consistency.
1 code implementation • 8 Apr 2022 • Weiguang Zhao, Chaolong Yang, Jianan Ye, Rui Zhang, Yuyao Yan, Xi Yang, Bin Dong, Amir Hussain, Kaizhu Huang
While weakly supervised multi-view face reconstruction (MVR) is garnering increased attention, one critical issue still remains open: how to effectively fuse multiple image information to reconstruct high-precision 3D models.
no code implementations • 26 Mar 2022 • Zhuang Qian, Kaizhu Huang, Qiu-Feng Wang, Xu-Yao Zhang
In this paper, we present a comprehensive survey trying to offer a systematic and structured investigation on robust adversarial training in pattern recognition.
no code implementations • 18 Feb 2022 • Chenru Jiang, Kaizhu Huang, Shufei Zhang, Jimin Xiao, Zhenxing Niu, Amir Hussain
In this paper, we focus on tackling the precise keypoint coordinates regression task.
1 code implementation • 27 Jan 2022 • Penglei Gao, Xi Yang, Rui Zhang, John Y. Goulermas, Yujie Geng, Yuyao Yan, Kaizhu Huang
In this paper, we develop a novel transformer-based generative adversarial neural network called U-Transformer for generalised image outpainting problem.
1 code implementation • 1 Nov 2021 • Kai Yao, Kaizhu Huang, Jie Sun, Amir Hussain
Automatic nuclei segmentation and classification play a vital role in digital pathology.
Ranked #4 on Multi-tissue Nucleus Segmentation on CoNSeP
1 code implementation • 21 Oct 2021 • Liuqing Zhao, Fan Lyu, Fuyuan Hu, Kaizhu Huang, Fenglei Xu, Linyan Li
Sentence-based Image Editing (SIE) aims to deploy natural language to edit an image.
no code implementations • 29 Sep 2021 • Zhuang Qian, Shufei Zhang, Kaizhu Huang, Qiufeng Wang, Bin Gu, Huan Xiong, Xinping Yi
It is possibly due to the fact that the conventional adversarial training methods generate adversarial perturbations usually in a supervised way, so that the adversarial samples are highly biased towards the decision boundary, resulting in an inhomogeneous data distribution.
no code implementations • 23 Jul 2021 • Kai Yao, Kaizhu Huang, Jie Sun, Curran Jude
We also propose a novel training algorithm able to align the disentangled content in the latent space to reduce micro-level lossy transformation.
1 code implementation • 8 Jul 2021 • Zhuang Qian, Shufei Zhang, Kaizhu Huang, Qiufeng Wang, Rui Zhang, Xinping Yi
The proposed adversarial training with latent distribution (ATLD) method defends against adversarial attacks by crafting LMAEs with the latent manifold in an unsupervised manner.
no code implementations • 16 Jun 2021 • Shuyi Qu, Zhenxing Niu, Kaizhu Huang, Jianke Zhu, Matan Protter, Gadi Zimerman, Yinghui Xu
Recent deep generative models have achieved promising performance in image inpainting.
1 code implementation • 16 Jun 2021 • Zihan Ye, Fuyuan Hu, Fan Lyu, Linyan Li, Kaizhu Huang
However, the traditional TL cannot search reliable unseen disentangled representations due to the unavailability of unseen classes in ZSL.
no code implementations • 13 Jun 2021 • Zhicheng Cai, Kaizhu Huang, Chenglei Peng
This paper proposes a novel nonlinear activation mechanism typically for convolutional neural network (CNN), named as reborn mechanism.
no code implementations • 28 Apr 2021 • Yangfan Zhou, Kaizhu Huang, Cheng Cheng, Xuguang Wang, Amir Hussain, Xin Liu
%on how to exploit strong convexity to further improve the convergence rate of AdaBelief.
1 code implementation • 24 Apr 2021 • Jingwei Guo, Kaizhu Huang, Xinping Yi, Rui Zhang
}, we introduce a novel Local and Global Disentangled Graph Convolutional Network (LGD-GCN) to capture both local and global information for graph disentanglement.
no code implementations • 10 Mar 2021 • Ping Guo, Kaizhu Huang, Zenglin Xu
In this work, we generalize the reaction-diffusion equation in statistical physics, Schr\"odinger equation in quantum mechanics, Helmholtz equation in paraxial optics into the neural partial differential equations (NPDE), which can be considered as the fundamental equations in the field of artificial intelligence research.
1 code implementation • ICCV 2021 • Zhiqiang Gao, Shufei Zhang, Kaizhu Huang, Qiufeng Wang, Chaoliang Zhong
In particular, we show that the distribution discrepancy can be reduced by constraining feature gradients of two domains to have similar distributions.
1 code implementation • 26 Nov 2020 • Penglei Gao, Xi Yang, Rui Zhang, Kaizhu Huang
We propose a continuous neural network architecture, termed Explainable Tensorized Neural Ordinary Differential Equations (ETN-ODE), for multi-step time series prediction at arbitrary time points.
2 code implementations • NeurIPS 2021 • Ye Ma, Zixun Lan, Lu Zong, Kaizhu Huang
A global scoring mechanism is then developed to regulate beam search to generate summaries in a near-global optimal fashion.
1 code implementation • 13 Dec 2018 • Haochuan Jiang, Guanyu Yang, Kaizhu Huang, and Rui ZHANG
Due to the huge category number, the sophisticated com-binations of various strokes and radicals, and the free writing or print-ing styles, generating Chinese characters with diverse styles is alwaysconsidered as a difficult task.
2 code implementations • 6 Dec 2017 • Kyeong Soo Kim, Sanghyuk Lee, Kaizhu Huang
Exploiting the hierarchical nature of the building/floor estimation and floor-level coordinates estimation of a location, we propose a new DNN architecture consisting of a stacked autoencoder for the reduction of feature space dimension and a feed-forward classifier for multi-label classification of building/floor/location, on which the multi-building and multi-floor indoor localization system based on Wi-Fi fingerprinting is built.
no code implementations • 15 Mar 2012 • Kaizhu Huang, Rong Jin, Zenglin Xu, Cheng-Lin Liu
Most existing distance metric learning methods assume perfect side information that is usually given in pairwise or triplet constraints.