Cognitive diagnostics in the Web-based Intelligent Education System (WIES) aims to assess students' mastery of knowledge concepts from heterogeneous, noisy interactions. Recent work has tried to utilize Large Language Models (LLMs) for cognitive diagnosis, yet LLMs struggle with structured data and are prone to noise-induced misjudgments. Specially, WIES's open environment continuously attracts new students and produces vast amounts of response logs, exacerbating the data imbalance and noise issues inherent in traditional educational systems. To address these challenges, we propose DLLM, a Diffusion-based LLM framework for noise-robust cognitive diagnosis. DLLM first constructs independent subgraphs based on response correctness, then applies relation augmentation alignment module to mitigate data imbalance. The two subgraph representations are then fused and aligned with LLM-derived, semantically augmented representations. Importantly, before each alignment step, DLLM employs a two-stage denoising diffusion module to eliminate intrinsic noise while assisting structural representation alignment. Specifically, unconditional denoising diffusion first removes erroneous information, followed by conditional denoising diffusion based on graph signal to eliminate misleading information. Finally, the noise-robust representation that integrates semantic knowledge and structural information is fed into existing cognitive diagnosis models for prediction. Experimental results on three publicly available web-based educational platform datasets demonstrate that our DLLM achieves optimal predictive performance across varying noise levels, which demonstrates that DLLM achieves noise robustness while effectively leveraging semantic knowledge from LLM.

With the widespread use of Graph Neural Networks (GNNs) for representation learning from network data, the fairness of GNN models has raised great attention lately. Fair GNNs aim to ensure that node representations can be accurately classified, but not easily associated with a specific group. Existing advanced approaches essentially enhance the generalisation of node representation in combination with data augmentation strategy and do not directly impose constraints on the fairness of GNNs. In this work, we identify that a fundamental reason for the unfairness of GNNs is the phenomenon of social homophily , i.e., users in the same group are more inclined to congregate. The message-passing mechanism of GNNs can cause users in the same group to have similar representations due to social homophily, leading model predictions to establish spurious correlations with sensitive attributes. Inspired by this reason, we propose a method called Equity-Aware GNN (EAGNN) towards fair graph representation learning. Specifically, to ensure that model predictions are independent of sensitive attributes while maintaining prediction performance, we introduce constraints for fair representation learning based on three principles: sufficiency, independence and separation. We theoretically demonstrate that our EAGNN method can effectively achieve group fairness. Extensive experiments on three datasets with varying levels of social homophily illustrate that our EAGNN method achieves the state-of-the-art performance across two fairness metrics and offers competitive effectiveness.

Anomaly detection in attributed graph data has been widely applied in real applications. However, the intricate topology of graph data, high-dimensional attributes, and class imbalance inherent in anomaly detection tasks render attributed graph anomaly detection a challenging task. To detect anomalies using the intricate topology information of graph data, a dual-masked autoencoders is proposed for attributed graph anomaly detection, denoted as MAGAD. Specifically, in the MAGAD, the class imbalance in attributed graph data is dealt with by randomly masking the original graph data to obtain masked graph data for the anomaly detection task. And then, a latent representation of the graph data is obtained by training dual autoencoders, where one autoencoder is developed for reconstructing the original graph data, and another for reconstructing randomly masked graph data. This assists in identifying abnormal nodes in the attributed graph data. Subsequently, to capture anomalous information from relevant features, MAGAD uses a random re-masking strategy for latent representations learned from the masked graph. Finally, the anomaly scores of the nodes are calculated using the learned latent representations from the decoders of the dual autoencoders. Experimental results on five real-world datasets demonstrate that the MAGAD algorithm outperforms state-of-the-art anomaly detection algorithms.

The rapid development of Large Language Models (LLMs) creates new opportunities for recommender systems, especially by exploiting the side information (e.g., descriptions and analyses of items) generated by these models. However, aligning this side information with collaborative information from historical interactions poses significant challenges. The inherent biases within LLMs can skew recommendations, resulting in distorted and potentially unfair user experiences. On the other hand, propensity bias causes side information to be aligned in such a way that it often tends to represent all inputs in a low-dimensional subspace, leading to a phenomenon known as dimensional collapse, which severely restricts the recommender system’s ability to capture user preferences and behaviors. To address these issues, we introduce a novel framework named Counterfactual LLM Recommendation (CLLMR). Specifically, we propose a spectrum-based side information encoder that implicitly embeds structural information from historical interactions into the side information representation, thereby circumventing the risk of dimension collapse. Furthermore, our CLLMR approach explores the causal relationships inherent in LLM-based recommender systems. By leveraging counterfactual inference, we counteract the biases introduced by LLMs. Extensive experiments demonstrate that our CLLMR approach consistently enhances the performance of various recommender models.

It has become a cardinal task to identify fake information (misinformation) on social media, because it has significantly harmed the government and the public. There are many spam bots maliciously retweeting misinformation. This study proposes an efficient model for detecting misinformation with self-supervised contrastive learning. A B ayesian graph L ocal extrema C onvolution (BLC) is first proposed to aggregate node features in the graph structure. The BLC approach considers unreliable relationships and uncertainties in the propagation structure, and the differences between nodes and neighboring nodes are emphasized in the attributes. Then, a new long-tail strategy for matching long-tail users with the global social network is advocated to avoid over-concentration on high-degree nodes in graph neural networks. Finally, the proposed model is experimentally evaluated with two public Twitter datasets and demonstrates that the proposed long-tail strategy significantly improves the effectiveness of existing graph-based methods in terms of detecting misinformation. The robustness of BLC has also been examined on three graph datasets and demonstrates that it consistently outperforms traditional algorithms when perturbed by 15% of a dataset.

Abstract Visual question answering is a complex multimodal task involving images and text, with broad application prospects in human–computer interaction and medical assistance. Therefore, how to deal with the feature interaction and multimodal feature fusion between the critical regions in the image and the keywords in the question is an important issue. To this end, we propose a neural network based on the encoder–decoder structure of the transformer architecture. Specifically, in the encoder, we use multi-head self-attention to mine word–word connections within question features and stack multiple layers of attention to obtain multi-level question features. We propose a mutual attention module to perform information exchange between modalities for better question features and image features representation on the decoder side. Besides, we connect the encoder and decoder in a meshed manner, perform mutual attention operations with multi-level question features, and aggregate information in an adaptive way. We propose a multi-scale fusion module in the fusion stage, which utilizes feature information at different scales to complete modal fusion. We test and validate the model effectiveness on VQA v1 and VQA v2 datasets. Our model achieves better results than state-of-the-art methods.