Traditional hate speech detection methods primarily focus on surface-level semantic cues and emotional intensity, resulting in suboptimal performance when identifying implicit and complex hate speech. To address this challenge, this study introduces cognitive distortion as an actionable reasoning cue and proposes a Cognitive Distortion-Aware Dual-Source Knowledge Distillation (CDA-DKD) framework. The framework employs a two-stage strategy. First, an Expert-Guided Cognitive Amplification (EGCA) mechanism is designed to enrich sparse signals and resolve ambiguous boundaries associated with cognitive distortions in training data. Second, a Dual-Teacher Single-Student Dual-Source Knowledge Distillation (DSKD) architecture is constructed. This architecture efficiently distills knowledge from two specialized expert teachers—one focusing on semantics and the other on cognitive distortions—into a lightweight student model. Evaluated on three public datasets (IHC, SBIC, and DYNA), the proposed framework achieves state-of-the-art Macro F1-scores of 79.73%, 91.34%, and 84.25%, respectively. Experimental results demonstrate that CDA-DKD significantly improves detection robustness, particularly in identifying implicit hate speech devoid of explicit offensive vocabulary.

Abstract Group‐oriented applications show its potential ability in the next generation of wireless sensor networks (5G WSNs), which have the particularity of being heterogeneous and so have different capabilities in terms of storage, computing, communicating and energy. One of the main challenges for secure group‐oriented applications (SGA) in 5G WSNs is how to secure communication between these heterogeneous devices. Conventional protocols are not suitable for SGA in 5G sensor networks since multiparty output establishment in this environment requires lightweight communication and computation overhead, further the primary task of SGA in 5G WSNs is to securely transmit various types of jointly computing data. Hence, membership authentication and multiparty output for arithmetic computations become two fundamental and necessary security services in SGA for 5G WSNs. In this paper we propose a novel design of non‐interactive integrated membership authenticated multiparty output for arithmetic computations in 5G sensor networks, which embeds the function of membership authentication and multiparty output for arithmetic computations. Since any arithmetic computation function is composed of multiple additions and multiplications, our result serves as a general method for multiparty computation output in SGA. This design is more suitable for lightweight membership authenticated multiparty arithmetic computations output in 5G sensor networks.

Semantic Knowledge is usually adding into topic model to improve topic coherence. However, it's hard to judge whether semantic information is related to topic without using complicated lexical characteristics. In this paper, we demonstrate a novel model called Cloud Transformation Model, which can easily judge whether semantic information is related to topic, and integrate semantic information into topic model.

This paper presents a text retrieval method based on vector space feature conversion. Even though knowledge-based semantic fingerprint information and semantic information obtained by the Tag-LDA topic model are two different representations of semantic features of the same text, they are incompatible with semantic information. Here we introduce the vector space as a bridge to transfer knowledge-based semantic fingerprint information space into the Tag-LDA model space and prove the rationality of the conversion process with correlation theory. The compatible semantic fingerprint information into the Tag-LDA model generate a new topic model STag-LDA. The Stag-LDA model has certain disambiguation effect for the semantic information of the tag. So, it can mine text semantic information more accurately to improve the retrieval efficiency.