Network structure evolves with time in the real world, and the discovery of changing communities in dynamic networks is an important research topic that poses challenging tasks. Most existing methods assume that no significant change in the network occurs; namely, the difference between adjacent snapshots is slight. However, great change exists in the real world usually. The great change in the network will result in the community detection algorithms are difficulty obtaining valuable information from the previous snapshot, leading to negative transfer for the next time steps. This paper focuses on dynamic community detection with substantial changes by integrating higher-order knowledge from the previous snapshots to aid the subsequent snapshots. Moreover, to improve search efficiency, a higher-order knowledge transfer strategy is designed to determine first-order and higher-order knowledge by detecting the similarity of the adjacency matrix of snapshots. In this way, our proposal can better keep the advantages of previous community detection results and transfer them to the next task. We conduct the experiments on four real-world networks, including the networks with great or minor changes. Experimental results in the low-similarity datasets demonstrate that higher-order knowledge is more valuable than first-order knowledge when the network changes significantly and keeps the advantage even if handling the high-similarity datasets. Our proposal can also guide other dynamic optimization problems with great changes.
Text retrieval is a long-standing research topic on information seeking, where a system is required to return relevant information resources to user's queries in natural language. From classic retrieval methods to learning-based ranking functions, the underlying retrieval models have been continually evolved with the ever-lasting technical innovation. To design effective retrieval models, a key point lies in how to learn the text representation and model the relevance matching. The recent success of pretrained language models (PLMs) sheds light on developing more capable text retrieval approaches by leveraging the excellent modeling capacity of PLMs. With powerful PLMs, we can effectively learn the representations of queries and texts in the latent representation space, and further construct the semantic matching function between the dense vectors for relevance modeling. Such a retrieval approach is referred to as dense retrieval, since it employs dense vectors (a.k.a., embeddings) to represent the texts. Considering the rapid progress on dense retrieval, in this survey, we systematically review the recent advances on PLM-based dense retrieval. Different from previous surveys on dense retrieval, we take a new perspective to organize the related work by four major aspects, including architecture, training, indexing and integration, and summarize the mainstream techniques for each aspect. We thoroughly survey the literature, and include 300+ related reference papers on dense retrieval. To support our survey, we create a website for providing useful resources, and release a code repertory and toolkit for implementing dense retrieval models. This survey aims to provide a comprehensive, practical reference focused on the major progress for dense text retrieval.
Video anomaly detection (VAD) has been intensively studied for years because of its potential applications in intelligent video systems. Existing unsupervised VAD methods tend to learn normality from training sets consisting of only normal videos and regard instances deviating from such normality as anomalies. However, they often consider only local or global normality. Some of them focus on learning local spatiotemporal representations from consecutive frames in video clips to enhance the representation for normal events. But powerful representation allows these methods to represent some anomalies and causes missed detections. In contrast, the other methods are devoted to memorizing global prototypical patterns of whole training videos to weaken the generalization for anomalies, which also restricts them to represent diverse normal patterns and causes false alarms. To this end, we propose a two-branch model, Local-Global Normality Network (LGN-Net), to learn local and global normality simultaneously. Specifically, one branch learns the evolution regularities of appearance and motion from consecutive frames as local normality utilizing a spatiotemporal prediction network, while the other branch memorizes prototype features of the whole videos as global normality by a memory module. LGN-Net achieves a balance of representing normal and abnormal instances by fusing local and global normality. The fused normality enables our model more generalized to various scenes compared to exploiting single normality. Experiments demonstrate the effectiveness and superior performance of our method. The code is available online: https://github.com/Myzhao1999/LGN-Net.
As an excellent tool for aiding communication, intelligent reflecting surface (IRS) can extend the coverage area, remove blind area, and achieve a dramatic rate improvement. In this paper, we improve the secret rate (SR) performance at directional modulation (DM) networks using IRS. To fully explore the benefits of IRS, two efficient methods are proposed to enhance SR performance. The first approach computes the confidential message (CM) beamforming vector by maximizing the SR, and the signal-to-leakage-noise ratio (SLNR) method is used to optimize the IRS phase shift matrix, which is called Max-SR-SLNR. Here, Eve is maximally interfered by transmiting artificial noise (AN) along the direct path and null-space projection (NSP) on the remaining two channels. To reduce the computational complexity, the CM, AN beamforming and IRS phase shift design are independently designed in the following methods. The CM beamforming vector is constructed based on maximum ratio transmission (MRT) criteria along the channel from Alice-to-IRS, and phase shift matrix of IRS is directly given by phase alignment (PA) method. This method is called MRT-NSP-PA. Simulation results show that the SR performance of the Max-SR-SLNR method outperforms the MRT-NSP-PA method in the cases of small-scale and medium-scale IRSs, and the latter approaches the former as IRS tends to lager-scale.
Stylized 3D avatars have become increasingly prominent in our modern life. Creating these avatars manually usually involves laborious selection and adjustment of continuous and discrete parameters and is time-consuming for average users. Self-supervised approaches to automatically create 3D avatars from user selfies promise high quality with little annotation cost but fall short in application to stylized avatars due to a large style domain gap. We propose a novel self-supervised learning framework to create high-quality stylized 3D avatars with a mix of continuous and discrete parameters. Our cascaded domain bridging framework first leverages a modified portrait stylization approach to translate input selfies into stylized avatar renderings as the targets for desired 3D avatars. Next, we find the best parameters of the avatars to match the stylized avatar renderings through a differentiable imitator we train to mimic the avatar graphics engine. To ensure we can effectively optimize the discrete parameters, we adopt a cascaded relaxation-and-search pipeline. We use a human preference study to evaluate how well our method preserves user identity compared to previous work as well as manual creation. Our results achieve much higher preference scores than previous work and close to those of manual creation. We also provide an ablation study to justify the design choices in our pipeline.
Image super-resolution is a common task on mobile and IoT devices, where one often needs to upscale and enhance low-resolution images and video frames. While numerous solutions have been proposed for this problem in the past, they are usually not compatible with low-power mobile NPUs having many computational and memory constraints. In this Mobile AI challenge, we address this problem and propose the participants to design an efficient quantized image super-resolution solution that can demonstrate a real-time performance on mobile NPUs. The participants were provided with the DIV2K dataset and trained INT8 models to do a high-quality 3X image upscaling. The runtime of all models was evaluated on the Synaptics VS680 Smart Home board with a dedicated edge NPU capable of accelerating quantized neural networks. All proposed solutions are fully compatible with the above NPU, demonstrating an up to 60 FPS rate when reconstructing Full HD resolution images. A detailed description of all models developed in the challenge is provided in this paper.
Federated clustering is an adaptation of centralized clustering in the federated settings, which aims to cluster data based on a global similarity measure while keeping all data local. The key here is how to construct a global similarity measure without sharing private data. To handle this, k-FED and federated fuzzy c-means (FFCM) respectively adapted K-means and fuzzy c-means to the federated learning settings, which aim to construct $K$ global cluster centroids by running K-means on a set of all local cluster centroids. However, the constructed global cluster centroids may be fragile and be sensitive to different non-independent and identically distributed (Non-IID) levels among clients. To handle this, we propose a simple but effective federated clustering framework with GAN-based data synthesis, which is called synthetic data aided federated clustering (SDA-FC). It outperforms k-FED and FFCM in terms of effectiveness and robustness, requires only one communication round, can run asynchronously, and can handle device failures. Moreover, although NMI is a far more commonly used metric than Kappa, empirical results indicate that Kappa is a more reliable one.
Multimodal representation learning has shown promising improvements on various vision-language tasks. Most existing methods excel at building global-level alignment between vision and language while lacking effective fine-grained image-text interaction. In this paper, we propose a jointly masked multimodal modeling method to learn fine-grained multimodal representations. Our method performs joint masking on image-text input and integrates both implicit and explicit targets for the masked signals to recover. The implicit target provides a unified and debiased objective for vision and language, where the model predicts latent multimodal representations of the unmasked input. The explicit target further enriches the multimodal representations by recovering high-level and semantically meaningful information: momentum visual features of image patches and concepts of word tokens. Through such a masked modeling process, our model not only learns fine-grained multimodal interaction, but also avoids the semantic gap between high-level representations and low- or mid-level prediction targets (e.g. image pixels), thus producing semantically rich multimodal representations that perform well on both zero-shot and fine-tuned settings. Our pre-trained model (named MAMO) achieves state-of-the-art performance on various downstream vision-language tasks, including image-text retrieval, visual question answering, visual reasoning, and weakly-supervised visual grounding.
Transformer is a transformative framework that models sequential data and has achieved remarkable performance on a wide range of tasks, but with high computational and energy cost. To improve its efficiency, a popular choice is to compress the models via binarization which constrains the floating-point values into binary ones to save resource consumption owing to cheap bitwise operations significantly. However, existing binarization methods only aim at minimizing the information loss for the input distribution statistically, while ignoring the pairwise similarity modeling at the core of the attention mechanism. To this end, we propose a new binarization paradigm customized to high-dimensional softmax attention via kernelized hashing, called EcoFormer, to map the original queries and keys into low-dimensional binary codes in Hamming space. The kernelized hash functions are learned to match the ground-truth similarity relations extracted from the attention map in a self-supervised way. Based on the equivalence between the inner product of binary codes and the Hamming distance as well as the associative property of matrix multiplication, we can approximate the attention in linear complexity by expressing it as a dot-product of binary codes. Moreover, the compact binary representations of queries and keys enable us to replace most of the expensive multiply-accumulate operations in attention with simple accumulations to save considerable on-chip energy footprint on edge devices. Extensive experiments on both vision and language tasks show that EcoFormer consistently achieves comparable performance with standard attentions while consuming much fewer resources. For example, based on PVTv2-B0 and ImageNet-1K, Ecoformer achieves a 73% energy footprint reduction with only a 0.33% performance drop compared to the standard attention. Code is available at https://github.com/ziplab/EcoFormer.
Vision Transformers (ViTs) have underpinned the recent breakthroughs in computer vision. However, designing the architectures of ViTs is laborious and heavily relies on expert knowledge. To automate the design process and incorporate deployment flexibility, one-shot neural architecture search decouples the supernet training and architecture specialization for diverse deployment scenarios. To cope with an enormous number of sub-networks in the supernet, existing methods treat all architectures equally important and randomly sample some of them in each update step during training. During architecture search, these methods focus on finding architectures on the Pareto frontier of performance and resource consumption, which forms a gap between training and deployment. In this paper, we devise a simple yet effective method, called FocusFormer, to bridge such a gap. To this end, we propose to learn an architecture sampler to assign higher sampling probabilities to those architectures on the Pareto frontier under different resource constraints during supernet training, making them sufficiently optimized and hence improving their performance. During specialization, we can directly use the well-trained architecture sampler to obtain accurate architectures satisfying the given resource constraint, which significantly improves the search efficiency. Extensive experiments on CIFAR-100 and ImageNet show that our FocusFormer is able to improve the performance of the searched architectures while significantly reducing the search cost. For example, on ImageNet, our FocusFormer-Ti with 1.4G FLOPs outperforms AutoFormer-Ti by 0.5% in terms of the Top-1 accuracy.