Self-supervised learning (SSL) has recently achieved great success in mining the user-item interactions for collaborative filtering. As a major paradigm, contrastive learning (CL) based SSL helps address data sparsity in Web platforms by contrasting the embeddings between raw and augmented data. However, existing CL-based methods mostly focus on contrasting in a batch-wise way, failing to exploit potential regularity in the feature dimension. This leads to redundant solutions during the representation learning of users and items. In this work, we investigate how to employ both batch-wise CL (BCL) and feature-wise CL (FCL) for recommendation. We theoretically analyze the relation between BCL and FCL, and find that combining BCL and FCL helps eliminate redundant solutions but never misses an optimal solution. We propose a dual contrastive learning recommendation framework---RecDCL. In RecDCL, the FCL objective is designed to eliminate redundant solutions on user-item positive pairs and to optimize the uniform distributions within users and items using a polynomial kernel for driving the representations to be orthogonal; The BCL objective is utilized to generate contrastive embeddings on output vectors for enhancing the robustness of the representations. Extensive experiments on four widely-used benchmarks and one industry dataset demonstrate that RecDCL can consistently outperform the state-of-the-art GNNs-based and SSL-based models (with an improvement of up to 5.65% in terms of Recall@20). The source code is publicly available https://github.com/THUDM/RecDCL

Clustering of multipath components (MPCs) is an important aspect of propagation channel modeling. When a time series of measurements, based on movement of transmitter and/or receiver, is available, the temporal evolution of MPCs can be used as a basis for clustering. We present an algorithm that bases clustering not only on distance of MPCs in the delay/angle space, but also how similar the temporal evolution of their parameters are. Sample results obtained from a vehicle-to-vehicle measurement campaign show good performance of the proposed algorithm.

Radio channel modeling has been an important research topic, since the performance of any communication system depends on channel characteristics. So far, most of existing clustering algorithms are conducted based on the multipath components (MPCs) extracted by using high-resolution parameter estimation approach, e.g., SAGE or MUSIC, etc. However, most of the estimation approaches require prior information to extract MPCs. Moreover, the high-resolution estimation approach usually results in relatively high complexity and thus the clusters can only be identified by using off-line approach after the measurements. Therefore, a power angle spectrum based clustering and tracking algorithm (PASCT) is proposed in this paper. Firstly, power angle spectrum (PAS) is extracted from measurement data by using Bartlett beamformer. For each PAS, potential targets are selected from background and separated into clusters by using image processing approaches. The recognized clusters are characterized into three attributes: i) size, ii) position, and iii) shape feature, where histogram of oriented is developed to describe the shape feature of the clusters. Moreover, a minimizing cost tracking approach based on Kuhn-Munkres method is proposed to accurately identify the clusters in non-stationary channels. The proposed PASCT algorithm is validated based on both simulations and measurements. It is found that the dominating clusters in both line-of-sight and non-line-of-sight environments can be well recognized and tracked with the proposed algorithm. By using the proposed algorithm, the dynamic changes of the clusters in real-time channel measurements, e.g., number, birthdeath process, and size of the clusters, can be well observed. Through the experiments, the proposed can achieve fairly good accuracy on the cluster identification with lower complexity compared to the conventional solution.

Malicious social bots achieve their malicious purposes by spreading misinformation and inciting social public opinion, seriously endangering social security, making their detection a critical concern. Recently, graph-based bot detection methods have achieved state-of-the-art (SOTA) performance. However, our research finds many isolated and poorly linked nodes in social networks, which graph-based methods cannot effectively detect. To address this problem, our research focuses on effectively utilizing node semantics and network structure to jointly detect sparsely linked nodes. Given the excellent performance of language models (LMs) in natural language understanding (NLU), we propose a novel social bot detection framework LGB, which consists of two main components: language model (LM) and graph neural network (GNN). Specifically, the social account information is first extracted into unified user textual sequences, which is then used to perform supervised fine-tuning (SFT) of the language model to improve its ability to understand social account semantics. Next, the semantically enriched node representation is fed into the pre-trained GNN to further enhance the node representation by aggregating information from neighbors. Finally, LGB fuses the information from both modalities to improve the detection performance of sparsely linked nodes. Extensive experiments on two real-world datasets demonstrate that LGB consistently outperforms state-of-the-art baseline models by up to 10.95%. LGB is already online: https://botdetection.aminer.cn/robotmain.

Social bot detection is becoming a task of wide concern in social security. All along, the development of social bot detection technology is hindered by the lack of high-quality annotated data. Besides, the rapid development of AI Generated Content (AIGC) technology is dramatically improving the creative ability of social bots. For example, the recently released ChatGPT [2] can fool the state-of-the-art AI-text-detection method with a probability of 74%, bringing a large challenge to content-based bot detection methods. To address the above drawbacks, we propose a Contrastive Learning-driven Social Bot Detection framework (CBD). The core of CBD is characterized by a two-stage model learning strategy: a contrastive pre-training stage to mine generalization patterns from massive unlabeled social graphs, followed by a semi-supervised fine-tuning stage to model task-specific knowledge latent in social graphs with a few annotations. The above strategy endows our model with promising detection performance under an extreme scarcity of labeled data. In terms of system architecture, we propose a smart feedback mechanism to further improve detection performance. Comprehensive experiments on a real bot detection dataset show that CBD consistently outperforms 10 state-of-the-art baselines by a large margin for few-shot bot detection using very little (5-shot) labeled data. CBD has been deployed online.