Detecting anomalous nodes in dynamic graphs is critical for applications such as social network analysis, financial risk management, and cybersecurity. Traditional methods often rely on Euclidean assumptions and linear operations to model temporal node behaviors, which can limit their effectiveness due to entangled norm and direction variations in node embeddings. To address these challenges, we propose SBF-Net (Spherical Behavior and Frequency-aware Network), a novel framework that projects node features onto a unit hypersphere to better disentangle norm and directional changes, enabling more accurate behavior quantification. To mitigate the scarcity of anomalous samples, we employ a Deep Belief Network-based anomaly synthesizer to generate diverse synthetic anomalies. Additionally, frequency domain analysis is incorporated to capture subtle and high-frequency anomaly patterns. Extensive experiments on multiple real-world dynamic graph datasets demonstrate that SBF-Net consistently improves anomaly detection performance over several strong baselines. For instance, on the Wikipedia dataset, SBF-Net achieves improvements of 6.32% in AUC and 7.93% in precision over existing methods.

Transformer-based detectors have advanced small-object detection, but they often remain inefficient and vulnerable to background-induced query noise, which motivates deep decoders to refine low-quality queries. We present HELP (Heatmap-guided Embedding Learning Paradigm), a noise-aware positional-semantic fusion framework that studies where to embed positional information by selectively preserving positional encodings in foreground-salient regions while suppressing background clutter. Within HELP, we introduce Heatmap-guided Positional Embedding (HPE) as the core embedding mechanism and visualize it with a heatbar for interpretable diagnosis and fine-tuning. HPE is integrated into both the encoder and decoder: it guides noise-suppressed feature encoding by injecting heatmap-aware positional encoding, and it enables high-quality query retrieval by filtering background-dominant embeddings via a gradient-based mask filter before decoding. To address feature sparsity in complex small targets, we integrate Linear-Snake Convolution to enrich retrieval-relevant representations. The gradient-based heatmap supervision is used during training only, incurring no additional gradient computation at inference. As a result, our design reduces decoder layers from eight to three and achieves a 59.4% parameter reduction (66.3M vs. 163M) while maintaining consistent accuracy gains under a reduced compute budget across benchmarks. Code Repository: https://github.com/yidimopozhibai/Noise-Suppressed-Query-Retrieval.

ABSTRACT Few‐shot relation extraction has attracted significant attention due to its potential to identify new relations when training samples are scarce. Previous studies demonstrate that meta‐learning techniques can significantly enhance models' adaptability in few‐shot scenarios. Most existing meta‐learning methods generalize to unseen relations by constructing effective prototype representations for each relation. However, meta‐learning techniques often require a large amount of labeled data during the meta‐training stage, which contradicts the initial purpose of addressing the issue of insufficient labeled data. Current methods usually fail to model both intra‐class similarity and inter‐class difference sufficiently, resulting in fuzzy class prototypes when distinguishing between similar but slightly different relations. Therefore, we propose a novel task, low‐resource few‐shot relation extraction (LR‐FSRE), that explores minimizing the usage of labeled data while maximizing the acquisition of task meta‐knowledge, and we design a novel graph‐based adaptive discrimination network that optimizes relation prototypes in both tasks. Specifically, we effectively represent class prototypes using the topological information between instance and relation‐level representation. Then, the relation discrimination network considers the difference between similar classes to further improve the recognition ability of relation classes. Finally, a debiased optimization strategy is employed to optimize the learning processes of task‐general and task‐specific knowledge. Experimental results show that our proposed framework outperforms the state‐of‐the‐art methods in FSRE and LR‐FSRE tasks, and achieves significant improvement in accuracy across challenging cross‐domain and similarity relation discrimination scenarios.

Opinion dynamics studies the principles governing the evolution of collective opinion, offering valuable insights into the comprehension of social phenomena and forecasting group behavior. However, existing opinion dynamics models often overlook the impact of both opinion climate and cognitive capacities on interactive behaviors, thus causing simulation outcomes diverge from real-world observations. Addressing this gap, we propose a novel opinion dynamics model based on hierarchical thinking to describe the opinion evolution on social networks. Individuals are classified into different levels according to their cognitive abilities. They act with bounded rationality at their respective levels to optimize both the promotion of personal opinions and the avoidance of cyberbullying. Through simulation analysis, we found the crucial role of users with high levels of hierarchical thinking. They can discern the opinion climate and articulate their opinion, acting as bridges in the evolution of public opinion. Their opinion can reach the bounded confidence range of more people, thereby enabling polarization to shift to consensus under the same conditions. Furthermore, this effect is independent of individual inherent attributes, which is more in line with real-life scenarios.

Closely monitoring service performance and detecting anomalies are critical for Internet-based services. However, even though dozens of anomaly detectors have been proposed over the years, deploying them to a given service remains a great challenge, requiring manually and iteratively tuning detector parameters and thresholds. This paper tackles this challenge through a novel approach based on supervised machine learning. With our proposed system, Opprentice (Operators' apprentice), operators' only manual work is to periodically label the anomalies in the performance data with a convenient tool. Multiple existing detectors are applied to the performance data in parallel to extract anomaly features. Then the features and the labels are used to train a random forest classifier to automatically select the appropriate detector-parameter combinations and the thresholds. For three different service KPIs in a top global search engine, Opprentice can automatically satisfy or approximate a reasonable accuracy preference (recall >= 0.66 and precision>= 0.66). More importantly, Opprentice allows operators to label data in only tens of minutes, while operators traditionally have to spend more than ten days selecting and tuning detectors, which may still turn out not to work in the end.

Page Views (PVs) are very crucial for search engines due to their close relationship to the revenue. When PVs change significantly, operators must be informed so that they can diagnose and fix the problem quickly, and prevent further loss. In reality, PVs can be counted in many ways (e.g., PVs originated from different ISPs), and different PVs are of different interest to operators (e.g., the PVs of a larger ISP is more important). As a result, different PVs often require different detection standards, or thresholds. However, attempts to tune a number of thresholds have been hampered by the cost of the manual effort involved. To address the above problem, we propose a practical framework, called PTL (practical threshold learning). Operators only need to provide a few simple labels about the detection results, then PTL will automatically tune the thresholds for different PVs. Using 4-month PVs from a global top search engine, our evaluation demonstrates that PTL can improve the accuracy of detection dramatically. More importantly, it introduces very little labeling overhead for operators. For example, when detecting the PVs of 103 ISPs, PTL can reduce the overall false negative rate from 96% to 9% using only 29 labels per week on average.