Few-shot relation extraction (FSRE) constitutes a critical task in natural language processing (NLP), involving learning relationship characteristics from limited instances to enable the accurate classification of new relations. The existing research primarily concentrates on using prototype networks for FSRE and enhancing their performance by incorporating external knowledge. However, these methods disregard the potential interactions among different prototype networks, and each prototype network can only learn and infer from its limited instances, which may limit the robustness and reliability of the prototype representations. To tackle the concerns outlined above, this paper introduces a novel prototype network called SACT (multi-head self-attention and contrastive-center loss), aimed at obtaining more comprehensive and precise interaction information from other prototype networks to bolster the reliability of the prototype network. Firstly, SACT employs a multi-head self-attention mechanism for capturing interaction information among different prototypes from traditional prototype networks, reducing the noise introduced by unknown categories with a small sample through information aggregation. Furthermore, SACT introduces a new loss function, the contrastive–center loss function, aimed at tightly clustering samples from a similar relationship category in the center of the feature space while dispersing samples from different relationship categories. Through extensive experiments on FSRE datasets, this paper demonstrates the outstanding performance of SACT, providing strong evidence for the effectiveness and practicality of SACT.

In the domain of password recovery, deep learning has emerged as a pivotal technology for enhancing recovery efficiency. Despite its effectiveness, the inherent computation complexity of deep learning-based password generation algorithms poses substantial challenges, particularly in achieving synergistic acceleration between deep learning inference, and plaintext encryption process. In this paper, we introduce PassRecover, a multi-FPGA-based computing system that can simultaneously accelerate deep learning-driven password generation and plaintext encryption in an end-to-end manner. The system architecture incorporates a neural processing unit (NPU) and an encryption array configured to operate under a streaming dataflow paradigm for parallel processing. It is the first approach to explore the benefit of end-to-end offline password recovery. For comprehensive evaluation, PassRecover is benchmarked against PassGAN and five industry-standard encryption algorithms (Office2010, Office2013, PDF1.7, Winzip, and RAR5). Experimental results demonstrate excellent performance: Compared to the latest work that only accelerate encryption algorithms, PassRecover achieves an average 101.5% speedup across all tested encryption algorithms. When compared to graphics processing unit (GPU)-based end-to-end implementations, this work delivers 93.01% faster processing speeds and $3.73\times$ superior energy efficiency. These results establish PassRecover as a promising solution for resource-constrained password recovery scenarios requiring high throughput and energy efficiency.

Self-supervised learning (SSL) is vulnerable to backdoor attacks, while the downstream classifiers based on SSL models inevitably inherit these backdoors, even when they are trained on clean samples. Despite the proposal of several methods of backdoor defense against backdoor attacks, few methods remain that can be used to effectively defend against various backdoor attacks while maintaining the high performance of the model. In this paper, based on the discovery that unlearning any trigger enhances the overall backdoor robustness of the model, a novel, efficient, and straightforward approach is proposed to address the most advanced backdoor attacks. This method involves two stages. Firstly, a backdoor is actively implanted in the model with a custom trigger. Secondly, the model is fine-tuned to unlearn the custom trigger. Through these two stages, it is found out that not only is the implanted backdoor removed, but the unknown backdoors implanted by attackers are also effectively mitigated. As illustrated by the extensive experiments conducted on multiple datasets against the current state-of-the-art methods of attack, the method proposed in this paper requires only a small amount of clean data (approximately 1%) to reduce the success rate of backdoor attacks effectively while ensuring minimal impact on model performance.

A fuzzy radial basis inference network with multimodality prior feature knowledge embedding is proposed for multimodality feature fusion and small sample set classification. This model mainly consists of a multi-channel modal feature input layer, a fuzzy radial basis neuron network (FRBN) layer, a modal category feature aggregation layer, a fuzzy rule layer, and a T-S fuzzy classifier. For the differences in spatiotemporal distribution, information granularity, and semantic representation of different modal features, FRBNs is used to embed multimodality diversity prior category feature knowledge, as well as to process the semantic and spatiotemporal information consistency of input modal features. Based on fuzzy computing logic, inference rules are established and multimodality features are fused layer by layer. Specifically, for classification tasks, high-order semantic features of each modality are extracted separately, and a similarity measurement function for modal features is constructed. Then, the fuzzy dynamic C-means clustering algorithm is used to select representative sample features from subsets of different categories in each modality, in order to implicitly represent prior category knowledge. They are used as the kernel center of FRBN to achieve the embedding of diversity prior feature knowledge. In the modality category feature aggregation layer, the diversity feature information of each modality category subclass is aggregated towards the category and generates a non-convex class interface. Fuzzy multiplication operation is adopted to establish fuzzy reasoning rules for classification tasks, and multimodality classification is implemented based on T-S fuzzy classifier. The proposed method can embed and utilize prior feature knowledge of each modality in a mechanism, impose structural and data constraints on the model, reduce the requirement for the completeness of the sample set, and maintain semantic consistency between modalities and within modalities. Applying the proposed method to medical diagnosis, four types of heart disease classification were performed using multimodality medical data such as echocardiogram, multi-lead electrocardiogram, myocardial enzyme examination, and clinical symptoms, with an accuracy rate of 83.37%. This verifies the effectiveness and application value of the proposed method.

Multimodal Large Language Models (MLLMs) have demonstrated remarkable capabilities across vision–language tasks, yet their safety alignment remains vulnerable to adversarial manipulation. Existing jailbreak attacks typically optimize adversarial perturbations using negative log-likelihood loss alone, which often leads to overfitting on target affirmative tokens and fails to elicit substantive harmful content. We propose Attention-Enhancement and Targeted Entropy Regularization for Adversarial Optimization (AERO), a novel jailbreak framework addressing these limitations through two complementary mechanisms. First, an attention enhancement loss strategically redirects cross-modal attention toward perturbed visual tokens, distracting safety-aligned features from scrutinizing malicious queries. Second, a targeted entropy regularization scheme maximizes output diversity over non-refusal tokens during initial generation, creating a permissive context that improves cross-query generalization and enables responses that genuinely address malicious requests. Extensive experiments on multiple state-of-the-art MLLMs demonstrate that AERO significantly outperforms existing methods, achieving Attack Success Rates (ASRs) of 65.8–70.7% on MM-SafetyBench and 71.0–84.5% on HarmBench. Our approach surpasses the strongest baselines by margins of up to 16.2% in success rate while consistently generating higher-quality harmful content.