Real-world graph data often follows long-tailed distributions, making it difficult for Graph Neural Networks (GNNs) to generalize well across both head and tail classes. Recent advances in Vicinal Risk Minimization (VRM) have shown promise in mitigating class imbalance with numeric interpolation; however, existing approaches largely rely on embedding-space arithmetic, which fails to capture the rich semantics inherent in text-attributed graphs. In this work, we propose our method, SaVe-TAG (Semantic-aware Vicinal Risk Minimization for Long-Tailed Text-Attributed Graphs), a novel VRM framework that leverages Large Language Models (LLMs) to perform text-level interpolation, generating on-manifold, boundary-enriching synthetic samples for minority classes. To mitigate the risk of noisy generation, we introduce a confidence-based edge assignment mechanism that uses graph topology as a natural filter to ensure structural consistency. We provide theoretical justification for our method and conduct extensive experiments on benchmark datasets, showing that our approach consistently outperforms both numeric interpolation and prior long-tailed node classification baselines. Our results highlight the importance of integrating semantic and structural signals for balanced and effective learning on text-attributed graphs. The source code is publicly available at: https://github.com/LWang-Laura/SaVe-TAG.

Recently, Knowledge Graphs (KGs) have been successfully coupled with Large Language Models (LLMs) to mitigate their hallucinations and enhance their reasoning capability, e.g., KG-based retrieval-augmented framework. However, current KG-LLM frameworks lack rigorous uncertainty estimation, limiting their reliable deployment in applications where the cost of errors is significant. Directly incorporating uncertainty quantification into KG-LLM frameworks presents a challenge due to their more complex architectures and the intricate interactions between the knowledge graph and language model components. To address this crucial gap, we propose a new trustworthy KG-LLM framework, UAG (Uncertainty Aware Knowledge-Graph Reasoning), which incorporates uncertainty quantification into the KG-LLM framework. We design an uncertainty-aware multi-step reasoning framework that leverages conformal prediction to provide a theoretical guarantee on the prediction set. To manage the error rate of the multi-step process, we additionally introduce an error rate control module to adjust the error rate within the individual components. Extensive experiments show that UAG can achieve any pre-defined coverage rate while reducing the prediction set/interval size by 40% on average over the baselines.

Fetal electrocardiography (FECG) presents an important avenue for continuous fetal monitoring. Nonetheless, effectively extracting FECG from maternal electrocardiogram (MECG) is one considerable challenge due to its weaker amplitude compared to MECG and the non-Gaussian nature of background noise. In this letter, we introduce alpha-stable noise to model the realistic interference due to its high scalability. To improve the accuracy of FECG extraction under impulsive noise (alpha-stable noise with strong impulses), we introduce the least Lncosh algorithm (Llncosh) and the improved Llncosh algorithm (ILL) is proposed based on the Amplitude Hyperbolic Tangent Transformation (AHTT) to optimize the preset parameter. Moreover, Monte Carlo experiments are carried out to investigate the capabilities of LMS-like algorithms and the ILL algorithm for FECG extraction. The results demonstrate that the ILL algorithm outperforms the LMS-like ones with carefully selected parameters, particularly showcasing superior robustness against impulsive noise. This work holds significance both in the theoretical research of adaptive filtering and in its practical application for FECG extraction.

Recently, Knowledge Graphs (KGs) have been successfully coupled with Large Language Models (LLMs) to mitigate their hallucinations and enhance their reasoning capability, e.g., KG-based retrieval-augmented framework for question-answering. However, current KG-LLM frameworks lack rigorous uncertainty estimation, limiting their reliable deployment in high-stake applications where the cost of errors is significant. To address this crucial gap, we propose a new trustworthy KG-LLM framework, UaG(Uncertainty Aware Graph Reasoning), which incorporates uncertainty quantification into the KG-LLM framework. We design an uncertainty-aware multi-step reasoning framework that leverages conformal prediction to provide a theoretical guarantee on the prediction set. To manage the error rate of the multi-step process, we additionally introduce an error rate control module to adjust the error rate within the individual components. Our preliminary results demonstrate that UaG can achieve the desired theoretical coverage while maintaining a reasonable prediction set size.

U-Net[1] framework which containing an encoder- decoder architecture is still a comment choice for semantic segmentation in medical area. However, due to the intrinsic locality of convolution operations, the U-Net framework is not capable of capturing long-range dependency. Transformer[2] which can model long-range dependency because of the insider self-attention mechanism, first proposed in natural language processing domain and got a great success, is introduced to computer vision and has achieved promising results in the downstream tasks such as image classification and segmentation. In this paper, we propose UTransNet to explore a way to fuse transformer into U-Net to take both advantage of the characteristics of convolution layer and transformer to segment medical images. We test our end-to-end network on ATLAS datasets and the results demonstrate that the performance of our method is superior than previous U-Net based methods but with the least parameters.

The proliferation of web platforms has created incentives for online abuse. Many graph-based anomaly detection techniques are proposed to identify the suspicious accounts and behaviors. However, most of them detect the anomalies once the users have performed many such behaviors. Their performance is substantially hindered when the users' observed data is limited at an early stage, which needs to be improved to minimize financial loss. In this work, we propose Eland, a novel framework that uses action sequence augmentation for early anomaly detection. Eland utilizes a sequence predictor to predict next actions of every user and exploits the mutual enhancement between action sequence augmentation and user-action graph anomaly detection. Experiments on three real-world datasets show that Eland improves the performance of a variety of graph-based anomaly detection methods. With Eland, anomaly detection performance at an earlier stage is better than non-augmented methods that need significantly more observed data by up to 15% on the Area under the ROC curve.

With the increasing popularity of portable devices with cameras (e.g., smartphones and tablets) and ubiquitous Internet connectivity, travelers can share their instant experience during the travel by posting photos they took to social media platforms. In this paper, we present a new image-driven poetry recommender system that takes a traveler's photo as input and recommends classical poems that can enrich the photo with aesthetically pleasing quotes from the poems. Three critical challenges exist to solve this new problem: i) how to extract the implicit artistic conception embedded in both poems and images? ii) How to identify the salient objects in the image without knowing the creator's intent? iii) How to accommodate the diverse user perceptions of the image and make a diversified poetry recommendation? The proposed iPoemRec system jointly addresses the above challenges by developing heterogeneous information network and neural embedding techniques. Evaluation results from real-world datasets and a user study demonstrate that our system can recommend highly relevant classical poems for a given photo and receive significantly higher user ratings compared to the state-of-the-art baselines.

In the feature modeling procedure, one-dimensional topological entities (edges) are always used as references or operational objects. Hence, to realize the integration of heterogeneous CAD systems, the corresponding one-dimensional topological entities must be found in the target CAD system to match the source ones. This paper presents a method to gain one-dimensional topological entity matching. The method is based on two algorithms, i.e. combining algorithm and matching algorithm. The combining algorithm is adopted to combine the edges retrieved in a source CAD system. Then, for each combined edge, the matching edges are found by using the matching algorithm in a target CAD system. The experiments prove that our method is valid for both offline and online integrations.