Diffusion models have demonstrated remarkable performance in speech synthesis, but typically require multi-step sampling, resulting in low inference efficiency. Recent studies address this issue by distilling diffusion models into consistency models, enabling efficient one-step generation. However, these approaches introduce additional training costs and rely heavily on the performance of pre-trained teacher models. In this paper, we propose ECTSpeech, a simple and effective one-step speech synthesis framework that, for the first time, incorporates the Easy Consistency Tuning (ECT) strategy into speech synthesis. By progressively tightening consistency constraints on a pre-trained diffusion model, ECTSpeech achieves high-quality one-step generation while significantly reducing training complexity. In addition, we design a multi-scale gate module (MSGate) to enhance the denoiser’s ability to fuse features at different scales. Experimental results on the LJSpeech dataset demonstrate that ECTSpeech achieves audio quality comparable to state-of-the-art methods under single-step sampling, while substantially reducing the model’s training cost and complexity.

The technique of recursive neighborhood aggregation has dominated the implementation of existing successful Graph Neural Networks (GNNs). However, the recursive information propagation across layers inevitably brings in extra calculations, potentially large variance, and difficulty of parallel computation. In this paper, we propose Graph Unfolding Networks (GUNets) as an alternative mechanism of recursive neighborhood aggregation for graph representation learning. Comparing to generic GNNs, our proposed GUNets are efficient, robust and practically effective. At their core, GUNets unfold the local structure of every node, i.e. the rooted tree, to a set of trajectories, and then adopt set function to capture the topology of the rooted subtree, which is more convenient for parallel computation than the recursive neighborhood aggregation process. More importantly, through a specific design of the set function, our architecture enables efficient and robust learning on large-scale graphs without resorting to any pruning of the rooted subtree that is usually necessary in generic GNNs. Extensive experiments on five large datasets (the number of nodes ranges from 104 to 106) show that our GUNets achieve comparable or even better results than current successful GNNs while gaining significantly more efficiency and lower accuracy variance. Codes can be found at github.com/GUNets/GUNets.

Multi-class object detection always involves the tasks of accurate target localization which is mainly related to bounding box regression. Smooth L1 loss is the most popular bounding box regression loss used in the current state-of-the-art object detection systems. However, such loss for regressing the parameters of a bounding box can't accurately and consistently regress the bounding box to the associated ground truth well. We instead propose the near-duplicated loss, a loss that better evaluate the disparity between the bounding box and the ground truth consistently. We present an approximate algorithm associated with a kernel function that not only considers the absolute distance but also involves the relative overlap area between the two bounding boxes. The new loss doesn't need additional supervision and is easy to embed into existing networks. Our final result, by incorporating the near-duplicated loss into the state-of-the-art object detection detectors (Faster RCNN, RetinaNet), shows consistent and significant improvements on popular object detection benchmarks (MS COCO and Pascal VOC).

Improving the performance of click-through rate (CTR) prediction remains one of the core tasks in online advertising systems. With the rise of deep learning, CTR prediction models with deep networks remarkably enhance model capacities. In deep CTR models, exploiting users' historical data is essential for learning users' behaviors and interests. As existing CTR prediction works neglect the importance of the temporal signals when embed users' historical clicking records, we propose a time-aware attention model which explicitly uses absolute temporal signals for expressing the users' periodic behaviors and relative temporal signals for expressing the temporal relation between items. Besides, we propose a regularized adversarial sampling strategy for negative sampling which eases the classification imbalance of CTR data and can make use of the strong guidance provided by the observed negative CTR samples. The adversarial sampling strategy significantly improves the training efficiency, and can be co-trained with the time-aware attention model seamlessly. Experiments are conducted on real-world CTR datasets from both in-station and out-station advertising places.

Subspace segmentation is the problem of segmenting (or grouping) a set of$n$data points into a number of clusters, with each cluster being a (linear) subspace. The recently established algorithms such as Sparse Subspace Clustering (SSC), Low-Rank Representation (LRR) and Low-Rank Subspace Segmentation (LRSS) are effective in terms of segmentation accuracy, but computationally inefficient as they possess a complexity of$O(n^{3})$, which is too high to afford for the case where$n$is very large. In this paper we devise a fast subspace segmentation algorithm with complexity of$O(n\log (n))$. This is achieved by firstly using partial Singular Value Decomposition (SVD) to approximate the solution of LRSS, secondly utilizing Locality Sensitive Hashing (LSH) to build a sparse affinity graph that encodes the subspace memberships, and finally adopting a fast Normalized Cut (NCut) algorithm to produce the final segmentation results. Besides of high efficiency, our algorithm also has comparable effectiveness as the original LRSS method.

Character rectification is very important for character recognition. Front view standard character images are much easier to recognize since most character recognition algorithms were trained with such data. However, the existing text rectification methods only work for a paragraph or a page. We discover that the modified TILT algorithm can be applied to rectify many single Chinese, English, and digit characters robustly. By changing the character image into a low-rank texture image via binarization and gray level inversion, the modified TILT method applies a rank minimization technique to recover the deformation and the proposed algorithm can work for almost all characters. To further enhance the robustness of the proposed algorithm, the modified TILT algorithm is extended for short phrases that consist of multiple characters. Extensive experiments testify to the effectiveness of the proposed method in rectifying texts with significant affine or perspective deformation in real images, such as street signs taken by mobile phones.

This article proposes a new approach to jointly rectify multi-distorted text image planes using a single image. Without extracting text lines or analyzing the text layout, the algorithm build a Multi-Distortion Dewarping (MDD) model based on modified text transform invariant low-rank textures. Harnessing the fact that two-intersection text plane share a same vanishing point, MDD algorithm greatly increase the estimation accuracy of geometry distortion. To further enhance the robustness of the propose method, a distorted text detection algorithm is used as pre-process to remove non-text region. With the accurately estimated geometry distortion of each plane, the input image can be well projected onto a single image plane and generate a good dewarping results. The MDD is robust to noise and works well for both short phrase and multiple text lines. Extensive compare experiments show the robustness and efficiency of MDD algorithm.

Circle ring aligned text is very common in our daily life, such as university logo, advertisement, there are quite few methods to expand the circle text which will greatly improve the working range of optical character recognition (OCR) product and the accuracy of text segmentation. In this paper, a new method is proposed to handle this circumstance, called curve Transform Invariant Low-rank Textures(TILT). By change the Cartesian system into polar system, the transformed image matrix D can be decomposed into low-rank matrix A and a sparse error matrix E. Matrix A represent the text expansion image and E is the noises and other non-regular component of text image. All this consist of an optimized convex problem and can be solved by alternating direction method (ADM) method. The proposed method also provides a frame work for curve text expansion. Extensive experiments show the robustness of proposed method in expanding artificial and real text image, which contain English or Chinese texts.