Various machine learning models have been employed to detect aggressive speech and cyberbullying incidents on social media platforms. Previous studies have not addressed the applicability of handling a lengthy social media session at a time due to the short content window and technical limitations imposed on early NLP models. In addition, some studies have focused on content classification without addressing the identification of related perpetrators and victims, which are important components for determining cyberbullying severity levels. This paper introduces a victim-perpetrator-category association modeling approach and investigates the practical use of recently developed long context length large language models to handle early cyberbullying incident detection tasks. Tailored prompt templates with proper stop phrases were designed to explicitly model the associations between victims and perpetrators, enabling a more accurate prediction of the occurrence of cyberbullying incidents. Two large pretrained models, Llama 3.1 8B and Qwen 2.5 14B, were finetuned using the low-rank adaptation fine-tuning technique with 1685 manually annotated multiparty Chinese-English Cantonese colloquial dialogue sessions, resulting in a total of 14,257 tweets. It is shown that the holistic use of dialogue session messages at a time provides significant performance advantages over handling messages individually in the victim identification task. The explicit victim-perpetrator-category association is empirically shown to improve early cyberbullying detection performance in terms of the ERDE and F-latency across all customizable severity levels gauged by different thresholds of insult frequency, number of perpetrators and power imbalance. • Explicit victim-perpetrator-category association for early bullying detection. • Identify cases by multi-thresholds: frequency, participants and power imbalance. • Demonstrate practical prompt templates for advanced generative LLMs. • Efficient fine-tuning architecture with rank stabilized low-rank adaptation method. • Annotate 1685 sessions of Chinese-English code-mixed colloquial language dialogues.

Background:: Bound-constrained single-objective optimization and constrained engineering design often feature heterogeneous landscapes and barrier-like structures, motivating search procedures that are scale-aware, robust near constraints, and economical in tuning. Contributions:: We introduce Quantum Tunneling and Fractional Calculus-Based Metaheuristic (QTFM), a physics-inspired metaheuristic that is parameter-lean and employs bounded, range-aware operators to reduce sensitivity to tuning and to prevent erratic steps close to constraints. Methodology:: QTFM couples fractional-step dynamics for scale-aware exploitation with a quantum-tunneling jump for barrier crossing, and augments these with a wavefunction-collapse local search that averages a small neighborhood and applies minimal perturbations to accelerate refinement without sacrificing diversity. Results:: On the IEEE Congress on Evolutionary Computation CEC 2022 single-objective bound-constrained suite, QTFM ranked first on ten of twelve functions; it reached the best optimum on F1 and achieved the best mean values on F2–F8 and F10–F11 with stable standard deviations. In three constrained engineering problems, QTFM produced the lowest mean and the best-found solution for the robotic gripper design, and the lowest mean for the planetary gear train and three-bar truss design. Findings:: The proposed fractional–quantum approach delivers fast, accurate, and robust search across heterogeneous landscapes and real-world design problems.