We present MSKE-LLM, a Multi-Stage Knowledge Enhancement (MSKE) policy question-answering Large Language Model (LLM). Its goal is to use structured policy knowledge graphs to enhance knowledge base retrieval ability, thereby improving the knowledge matching accuracy and response generation performance of LLMs when doing policy question-answering, a task to provide answers to questions regarding policies and regulations according to the relevant knowledge. Specifically, in response to the general inadequacy of knowledge in LLMs within the policy domain, and the decrease in retrieval accuracy as knowledge base size increases, we gather policy documents from local policy websites. Subsequently, we build an expansive policy knowledge graph to facilitate high-precision policy knowledge base matching, elevating knowledge retrieval accuracy and answer quality. In addition, we propose a standard for evaluating multi-stage knowledge enhancement policy question-answering LLM and conduct multi-dimensional human and automated evaluation and quantitative ablation research. The experimental results show that the proposed MSKE-LLM is superior to the existing models.

The widespread adoption of face recognition has raised privacy concerns regarding the collection and use of facial data. To address this, researchers have explored "unlearnable examples" by adding imperceptible perturbations during model training to prevent the model from learning target features. However, current methods are inefficient and cannot guarantee transferability and robustness at the same time, causing impracticality in the real world. To remedy it, we introduce Side-information Guided Generative Unlearnable Examples (Segue). Using a once-trained multiple-used model to generate perturbations, Segue avoids the time-consuming gradient-based approach. To improve transferability, we introduce side information such as true or pseudo labels, which are inherently consistent across different scenarios. For robustness enhancement, a distortion layer is integrated into the training pipeline. Experiments show Segue is 1000× faster than previous methods, transferable across datasets and models, and resistant to JPEG compression, adversarial training, and standard augmentations.

Naive retrieval-augmented generation (RAG) methods enhance large language models (LLMs) by retrieving relevant textual information, improving the accuracy of responses. However, they are limited in capturing the complex relationships and structures in textual graphs, where both textual and topological information are crucial for graph reasoning. To address this issue, we propose a multi-level pruning graph RAG framework, called MGRAG. MGRAG consists of three stages: sub-graph retrieval, multi-level graph pruning, and answer generation. We first index and rank the subgraphs to improve retrieval efficiency. Then, we apply dynamic pruning at both the subgraph and node levels to extract the most relevant graph structures. Finally, we integrate the query, graph description, and optimized graph structure as inputs to the LLM for response generation. Experimental results on multi-hop reasoning benchmarks demonstrate that MGRAG effectively eliminates irrelevant structures, significantly improving response accuracy and overall model performance.

The increasing realism of AI-generated videos has raised potential security concerns, making it difficult for humans to distinguish them from the naked eye. Despite these concerns, limited research has been dedicated to detecting such videos effectively. To this end, we propose an open-source AI-generated video detection dataset. Our dataset spans diverse objects, scenes, behaviors, and actions by organizing input prompts into independent dimensions. It also includes various generation models with different generative models, featuring popular commercial models such as OpenAI’s Sora, Google’s Veo, and Kwai’s Kling. Furthermore, we propose a simple yet effective Detection model based on Concistency of Frame (DeCoF), which learns robust temporal artifacts across different generation methods. Extensive experiments demonstrate the generality and efficacy of the proposed DeCoF in detecting AI-generated videos, including those from nowadays’ mainstream commercial generators.

Detecting deepfakes has been an increasingly important topic, especially given the rapid development of AI generation techniques. In this paper, we ask: How can we build a universal detection framework that is effective for most facial deepfakes? One significant challenge is the wide variety of deepfake generators available, resulting in varying forgery artifacts (e.g., lighting inconsistency, color mismatch, etc). But should we ``teach" the detector to learn all these artifacts separately? It is impossible and impractical to elaborate on them all. So the core idea is to pinpoint the more common and general artifacts across different deepfakes. Accordingly, we categorize deepfake artifacts into two distinct yet complementary types: Face Inconsistency Artifacts (FIA) and Up-Sampling Artifacts (USA). FIA arise from the challenge of generating all intricate details, inevitably causing inconsistencies between the complex facial features and relatively uniform surrounding areas. USA, on the other hand, are the inevitable traces left by the generator's decoder during the up-sampling process. This categorization stems from the observation that all existing deepfakes typically exhibit one or both of these artifacts. To achieve this, we propose a new data-level pseudo-fake creation framework that constructs fake samples with only the FIA and USA, without introducing extra less-general artifacts. Specifically, we employ a super-resolution to simulate the USA, while utilise image-level self-blending on diverse facial regions to create the FIA. We surprisingly found that, with this intuitive design, a standard image classifier trained only with our pseudo-fake data can non-trivially generalize well to previously unseen deepfakes.

Instant messaging with texts and stickers has become a widely adopted communication medium, enabling efficient expression of user semantics and emotions. With the increased use of stickers conveying information and feelings, sticker retrieval and recommendation has emerged as an important area of research. However, a major limitation in existing literature has been the lack of datasets capturing temporal and user-specific sticker interactions, which has hindered further progress in user modeling and sticker personalization. To address this, we introduce User-Sticker, a dataset that includes temporal and user anonymous ID across conversations. It is the largest publicly available sticker dataset to date, containing 22K unique users, 370K stickers, and 8.3M messages. The raw data was collected from a popular messaging platform from 67 conversations over 720 hours of crawling. All text and image data were carefully vetted for safety and privacy checks and modifications. Spanning 10 domains, the U-Sticker dataset captures rich temporal, multilingual, and cross-domain behaviors not previously available in other datasets. Extensive quantitative and qualitative experiments demonstrate U-Sticker's practical applications in user behavior modeling and personalized recommendation and highlight its potential to further research areas in personalized retrieval and conversational studies. U-Sticker dataset is publicly available.

Aspect-based sentiment analysis (ABSA) is dedicated to forecasting the sentiment polarity of aspect terms within sentences. Employing graph neural networks to capture structural patterns from syntactic dependency parsing has been confirmed as an effective approach for boosting ABSA. In most works, the topology of dependency trees or dependency-based attention coefficients is often loosely regarded as edges between aspects and opinions, which can result in insufficient and ambiguous syntactic utilization. To address these problems, we propose a new reinforced dependency graph convolutional network (RDGCN) that improves the importance calculation of dependencies in both distance and type views. Initially, we propose an importance calculation criterion for the minimum distances over dependency trees. Under the criterion, we design a distance-importance function that leverages reinforcement learning for weight distribution search and dissimilarity control. Since dependency types often do not have explicit syntax like tree distances, we use global attention and mask mechanisms to design type-importance functions. Finally, we merge these weights and implement feature aggregation and classification. Comprehensive experiments show the effectiveness of the criterion and importance functions. RDGCN yields excellent analysis results.

Poisoning efficiency is crucial in poisoning-based backdoor attacks, as attackers aim to minimize the number of poisoning samples while maximizing attack efficacy. Recent studies have sought to enhance poisoning efficiency by selecting effective samples. However, these studies typically rely on a proxy backdoor injection task to identify an efficient set of poisoning samples. This proxy attack-based approach can lead to performance degradation if the proxy attack settings differ from those of the actual victims, due to the shortcut nature of backdoor learning. Furthermore, proxy attack-based methods are extremely time-consuming, as they require numerous complete backdoor injection processes for sample selection. To address these concerns, we present a Proxy attack-Free Strategy (PFS) designed to identify efficient poisoning samples based on the similarity between clean samples and their corresponding poisoning samples, as well as the diversity of the poisoning set. The proposed PFS is motivated by the observation that selecting samples with high similarity between clean and corresponding poisoning samples results in significantly higher attack success rates compared to using samples with low similarity. Additionally, we provide theoretical foundations to explain the proposed PFS. We comprehensively evaluate the proposed strategy across various datasets, triggers, poisoning rates, architectures, and training hyperparameters. Our experimental results demonstrate that PFS enhances backdoor attack efficiency while also offering a remarkable speed advantage over previous proxy attack-based selection methodologies.

Most existing methods that address out-of-distribution (OOD) generalization for node classification on graphs primarily focus on a specific type of data biases, such as label selection bias or structural bias. However, anticipating the type of bias in advance is extremely challenging, and designing models solely for one specific type may not necessarily improve overall generalization performance. Moreover, limited research has focused on the impact of mixed biases, which are more prevalent and demanding in real-world scenarios. To address these limitations, we propose a novel Causality and Independence Enhancement (CIE) framework, applicable to various graph neural networks (GNNs). Our approach estimates causal and spurious features at the node representation level and mitigates the influence of spurious correlations through the backdoor adjustment. Meanwhile, independence constraint is introduced to improve the discriminability and stability of causal and spurious features in complex biased environments. Essentially, CIE eliminates different types of data biases from a unified perspective, without the need to design separate methods for each bias as before. To evaluate the performance under specific types of data biases, mixed biases, and low-resource scenarios, we conducted comprehensive experiments on five publicly available datasets. Experimental results demonstrate that our approach CIE not only significantly enhances the performance of GNNs but outperforms state-of-the-art debiased node classification methods.

This study aims to establish a comprehensive knowledge graph that spans domains and networks, with a specific focus on legal cases and their applications. The proposed methodology enables efficient collection and storage of large volumes of structured, semi-structured, and unstructured data related to cases from various sources including organizations, the government, and the internet. To analyze the relationships between roles in cases, a multimodal model is proposed to process and collect data for domain-specific knowledge graphs. Furthermore, to support social governance and public safety, a knowledge-driven intelligent recommendation algorithm is proposed in the form of question-answering, providing multiple strategies such as causal analysis, similar case matching and pre-disaster response. This work contributes to the field of artificial intelligence and natural language processing, with potential applications in legal and governmental domains, as well as in disaster response and prevention.